BOSTON UNIVE'RSITY
GRADUATE SCHOOL

Thesis

Bacterial Variation with Special Reference
to the Pneumococcus

by

Howard S. Friea.man

(B.S., College of the City of New York, 1941)

. submitted in uartial fulfilment of the
requirements for the degree of
Master of Arts
1947

378.T44-
r 3C 0

F\ n \MI

Anuroved

bv

Professor of Bacteriology ^

^ ^

I'irst Reader

Second Reader. . . . Jllftr

Instructor in Bacteriology

Digitized by the Internet Archive
in 2016 with funding from
Boston Library Consortium Member Libraries

https://archive.org/details/bacterialvariatiOOfrie

TOPICAL OUTLINE

Introduction- ------------------- -page 4

Bacterial Variability ------------------9

Pleomorphism and Monomorphism ------------- -12

Colonial Variation- ------------------ -19

Variations of the Pneumococcus- ------------ -27

Transformation of Pneumococcal Types- --------- -34

Nucleoproteins and Bacterial Genetics --------- -46

Abstract- ----------------------- -58

Bibliography- --------------------- -63

-

4

INTRODUCTION

during the latter half of the nineteenth century, which
saw the rise of scientific bacteriology and immunity, the sub-
jects under investigation in these sciences were unicellular
and multicellular microscopic organisms. In common with all
embryonic sciences, the major portion of these early investiga-
tions was directed towards the classification of those organ-
isms in either the plant or animal Kingdom. The first accepted
view7 was that thev belonged in the latter, and as such they
were referred to as "animalculi, " little animals. Eov/ever,
further investigations of such phenomena as cell walls , plasmo-
lysis, and various morphological characteristics led eventually
to their nresent classification, in the olant kingdom.

Ferdinand Cohn asserted, "They form the boundary line of
life; bevond them, life does not exist, so far at least as our
microscopic expedients reach; and these are not small." Indeed,
the microsconic size of bacteria led the nhysicist to consider
them as simple colloidal systems, while the chemist thought of
them as "bags of enzymes." This assumed primitiveness was not
confirmed biochemically until the end of the century when Winod-
gradskv ( 1887 ) announced that certain microorganisms were capable
of synthesizing their own protoplasm from mineral salts and
carbon dioxide. Here was a phenomenon, the svnthesis of highlv
complex organic substances from simple inorganic salts, vliich
might well be interpreted , in view of their occurrence in living

9

— V ..

,

,

,

5

organisms, as the beginning of life on earth. (Dubos )

Further cytological studies during the next fifty years
have revealed a complexity of structure in these "primitive”
organisms which is comparable to that found in the higher plant
and animal forms. They have further brought about the possibi-
litv of an anatomical comparison between these bacterial cells
and those which together make up the highly complex structures
of higher plants and animals. The processes which take place
within the cells of both types, assuming that v/e mav draw a line
of phylogenetic differentiation between them, have been found to
be similar in a great many respects. Although we are at present
incapable of determining the exact nature of many of these meta-
bolic phenomena occurring within various living cells, we are,
nevertheless, bv means of carefully controlled experimental
studies of various substances which we may identify within the
cell, and as belonging to a particular cell, able to draw cert-
ain conclusions concerning these metabolic nrocesses, by analogy
with certain similar phenomena which have occurred naturally _in
vitro or in vivo, or which have been made to occur in vitro or
in vivo by purely artificial and synthetic, or verv nearly arti-
ficial and synthetic means. Furthermore, it has been shown
McCarty, Avery, Feulgen, and others) that bacterial cells con-
tain writhin themselves certain chemical substances which are
known to be fundamental, not only to bacterial genera but to
cells of higher plants and animals, without which these cells
cannot carry on their life cycles. Thev have been shown to re-
quire certain essential substances for growth, these substances

.

.

.

6

being identical with the vitamins required for normal animal and
plant growth. They have been shown to require certain amino
acids as a minimum essential for survival, further, and by far
the most important analogy which has been drawn so far, is the
observation that many of the properties of bacterial cells are
identifiable with similar properties, not only of viruses but
also of the chromosomes and genes of plant and animal cells.

As we consider histology and microscopic anatomy to be the
study of hundreds of different types of cells, with many indiv-
idual characteristics and functions, analogously we mav consider
bacteriology to be the study of many different types of cells.

But whereas it is almost impossible, except under very carefully
controlled experimental conditions, to isolate any single animal
or plant cell from its usual environment and cause it to continue
to grow and exhibit its normal characteristics outside of the
animal organism, on the other hand it is possible to study one
particular bacterial cell, in reference to its growTth, means of
fission, and its various biochemical, Physical, morphological
and other properties. It is moreover possible to observe in a
short time the effect of an environmental change in an evolution -
ary phase of growth on a particular type of cell, due to its
rapid growth and frequent generation. Thus it may be possible
to study the effect of varied oxygen supply upon a culture of a
bacterium, e.g., staphylococcus aureus , which has been grown for
a long time in a controlled oxygen environment. It is then
bossible to observe the Properties of subsequent generations of
this bacterium, and so study the results of environmental adapt-

.

. . .

.

7

1

ivity, not over a oeriod of a thousand years but in a few days.

Further, as it has been found that certain bacterial and
animal cells have substances in common, it is possible to infer
the metabolism of the animal cells, and also the role of these
apparently fundamental comnonents of the cell, bv means of in
vitro experiments with the bacterial cells. One of the common
properties of bacterial and animal cells is that of producing a
mutation-like phenomenon at the rate of 1 per 10^ or 10^ cell
divisions. In the animal, v/hen this variant occurs, it general-
ly finds conditions unfavorable for its growth and continued ex-
istence, therefore it passes out of the anatomical picture. In
the case of the variant bacterial cell, it is possible to isolate
such cells when they occur, and to examine their differential
properties as compared with the parent strain, and with other
bacteria. Although it is not possible, except with gamma rays,
etc., to increase the incidence of the occurrence of these muta-
tions, it is nevertheless nossible to increase the rate of sur-
vival of these mutations by growing the parent strain on a mediu]
which will favor the growth of the variant, while at the same
time su 'inlying the parent strain with sufficient environmental
factors necessary for its grov/th. Depending unon the factors
that have caused the variation, and upon the type of variation,
there ma# or may not be a reversion to the parent strain.

It is the purpose of this paper to uresent one aspect of
the possibilities of aonlying the methods and investigations of
bacterial variation and transmutation. Transmutation enters
into the analysis to a greater extent than variation, the former

>

1

.

• •

.

.

8

term implying a variation which is transmissible from one gener-*
ation to the next, in series. Our intent is to anplv these
methods of investigation to the problems of the mechanisms of
metabolic processes within the animal cell. This side of the
nroblem has been apnroached from a purely bacteriological and
immunological point of view by many investigators, Alloway,

Aver-”-, Dawson, Dochez, Dubos , Griffith, Keidelberger , McCarty,
Hiemann, and others, through the studv of experiments connected
with pneumococcal variations, both permanent and temporary.

Before proceeding to this topic, it is necessary to point
out the various aspects of bacterial variability, as seen in
mutations exhibited bT' species other than Dinlococcus pneumoniae .
This will be the purpose of the first section of this pacer,
namely, an analysis of bacterial variants, both naturally
occuring and artificially induced. By this means we shall be
able to set down a groundwork of definitions and illustrations
from which we can better proceed to the problems of pneumococcal
transformation .

.

*

*

'

.

9

BACTilRIAl. variability

Bacteriology is a statistical science. It is not the study
of individual organisms; all the facts which together make up
that body of knowledge which we call bacteriology are derived
from the study of a great many generations of a bacterial type,
represented bv many millions of organisms. By subjecting this
large number of similar organisms to the identical experimental
conditions, which is the limit of our experimental methods at
present, we are able to observe the properties of a large number
of the organisms present. Thus our statistical data are drawn
from negative as well as positive results. For example, when we
sav that the organism iischerichia coli has the property of fer-
menting lactose, we base this conclusion upon the visual evidence
that when a colony of this organism has been isolated and trans-
planted to a tube of phenol red broth containing lactose, we ob-
serve a color change from red to vellow, indicating that acid
has been formed in the tube; v/e also observe gas in the collec-
tion tube. Statistically we may sajf that Bscherichia coli is
capable of fermenting lactose. However we know that in the fer-
mentation tube there are many millions of organisms present,
some of which are obviously capable of fermenting the carbohydr-
ate present. But we cannot say that every organism present has
that property. No one has yet devised a bacteriological tech-
nique wherteby the fermentative properties of a single cell of
the species Lscherichia coli , or of any other bacterial species
can be tested.

.

. ,

.

.

.

10

On this basis we might also sav that the majority of the
cells of Escherichia coli are incapable of fermenting lactose,
but that a sufficient number of naturally occurring variants
possessing that property are produced in the normal growth curve
of the organism during the 24-hour incubation period to make the

f

phenomenon visible. On the other hand, we have the organism
Escherichia, coli mutabile , which will not ferment lactose. In
this case the statistical evidence is on the negative side. We
can assume that there are not present sufficiently great numbers
of naturally occurring variants which are capable of fermenting
lactose to give visual evidence of that fact.

It appears that all bacterial cells are capable of produc-
ing mutant forms during their normal growth cycles. These vari-
ants ma^ or mar not be permanent . (Deskowitz ) Them may transmit
the >roperties i hich bear witness to their variance from the
parent strain, or thev mav revert to the parent strain, or thev
may give rise to both the narent strain of the variant and to
the variant itself. These naturally occurring variants cannot
be artificially controlled, but ther seem, according to Dubos ,
to occur regularly within the life cycle, and can therefore be
predicted with sur >rising accuracy.

The mutabile strain of Escherichia coli which was mentions .
before will usually revert after a time to the lactose-ferment-
ing form. Similar!^ the communis strain of the seme organism,
which does not ferment sucrose, gives rise to the conmunior
strain, which does ferment this sugar, and bv a stud^ of these ;

wo

strains and their alterations, the mutation from one to the oth ?r

.

.

.

'

.

11

may be adequately predicted.

Aside from these naturally occurring mutations, we are
mainly concerned with those variants which are artificially pro-
duced and which are generally stable and permanent. From a study-
of these mutations, effected by physiological, biochemical,
structural, serological and antigenic methods of variation, we
have been able to obtain a groundwork of knowledge which can be
anplied in the field of medical bacteriology to epidemiological
studies of various diseases, anu to the cytological and histo-
logical studies of higher plant and animal cells and of their
metabolic nrocesses and their structure.

►

.

12

PLEOMORFHISM AND MONOMORPHISM

In the early days of bacteriology one of the first manifest-
ations of bacterial variability to be observed and studied v/as
that of oleomorphism and monomorphism. It was thought that ther 5
was onlv a handful of bacterial types, and that these exhibited
marked variation in their morphology, and in their biochemical
properties . This early claim was eventually displaced, and its
rise to prominence w as attributed to faulty techniques, parti-
cularly in the manner of obtaining and maintaining cure cultures
of bacteria. It was then seen that the bacteria presented a
great multiplicity of distinct types.

The view which is generally accepted by present-dav bacter-
iologists is that the bacteria represent a phase of degeneration
°f the higher organisms, i.e., they have through evolution and
environmental adaptivity lost certain properties which alone
serve to distinguish them from the higher lants, particularly
the blue-green algae and the true or green algae, and so on up
the phylogenetic tree. It is possible in a general ay to
classify the bacteria not onl'1'- according to the properties which
in the main each genus or species may exhibit, but also to class-
ify them by the properties which thev may lack, and which serve
to differentiate them from each other, and also from the higher
plant organisms.

We find that the majority of the bacteria v/ill retain their
form from generation to generation, depending upon the culture
medium which is used, and aside from slight variations which may

13

be compared in man to the fact that some are tall or short, and
some are fat and thin, though their environment has not varied
sufficiently to warrant an explanation of these deviations by
reason of a degeneration or an evolution of the species. Howeve:
we do find certain bacteria which when grown on a suitable sub-
strate will, upon microscopical examination, exhibit varying
forms. This may be shorn by growing a single-cell culture, the
technique of which is very well defined! Avery , R.C. and Leland,
S.J.), to eliminate the possibility of contamination. One ex-
ample of this form of variation, which would seem to come under
the headinv of naturally occurring mutants, is Pasteurella tul-
arensis. This organism is commonlv found in the smear in both
coccoid and bacillary forms. In young cultures both forms may
be found, the coccoid form predominating. The coccoid form is
generally found alone in older cultures . This would appear to
indicate that the bacillary form is the variant . (Zinnser) The
mechanism of this mutation may possibly be explained in the
following manner. It is generally knowm that bacillary forms
tend to elongate due to an internal axially disposed force which
tends to counteract the rounding effect of surface tension on
the cells . (Dubos ) On this basis we may assume that the surface
tension of the culture medium generally used for the isolation
and cultivation of Pasteurella tularensis, a cystine-glucose
broth, will be sufficiently low to cause the naturally occurring
mutant forms to survive at a rate sufficient to be detected by
a microscopic smear. We must then further assume that unon agin
the surface tension og the medium is sufficiently increased, due

5

l- / .

♦

14

to the accumulation of secreted products from the organisms so
that only the coccoid f^rms are found. However, this hypothesis
cannot in any way be validified upon observation of cultures of
this organism grown uoon agar slants made from the same essentia,,
ingredients, since these cells also exhibit coccoid and bacillar; r
forms .

tie can only sav in this particular case that the mutation
is a natural phase of the cytomor ohosis of the cell during its
normal growth curve. This would appear to be the more valid
explanation.

It appears that many bacteria exhibit slight, but occasion-
ally marked variations in shape during their normal growth curve ,
These variations consist of differences in the size and contour
of the bacterial cell, the loss of certain structures, such as
capsules and flagella, variations in normal grouoing, and varia-
tions in the internal structure of the cell. Some of these nat-
urally occurring variations are illustrated below, along with
other variations which are interdenendent upon each other. This
interdependence of variations will be discussed later on, indeed
throughout the paper, for it appeals from all the data in the
literature that a single variation within a species is almost
unknown .

The streptococci generally show a v/ide variation in size at
practically all points along their normal growth curve, the dia-
meter ranging from 0.4 to 1.0 micron; these differences may be
observed within a single chain, large and small cocci being ad-
jacent to one another, showing that the one was the result of

.

.

,

,

.

.

the fission of the other. Thus we can differentiate this fact
from a possible observation of two distinct chains of cocci, in
each of which the cocci are of equal size, while the diameter
) varies from one chain to the other.

The Pneumococcus Type III may be found in the normal diplo-
coccal arrangement of the genus, but almost invariably exhibits
chain formation, which makes it difficult to distinguish the org-
anism from the streptococcus . This chain formation in turn is
generally associated with a variation in the shape of the organ-
ism. Instead of the normally occurring lanceolateo form, pneumo-
coccus Type III tends to be ovoid in shape, resembling not only
the streptococcus, but also the genus Neisseria, from v:hich, for-
tunately, we can as a rule differentiate it by means of the gram
stain. Photomicrographs taken of pneumococcus Type III colonies
on blood agar plates show this chain formation at the peri >hery
of the colony; the pneumococci are seen to be almost spherical in
^hape . (Bisset )

The genus Corynebacterium is. the most striking example of
pleomorphism. The forms vary from the evenly staining slender
bacilli, through barren ana granular varieties of clubbed and
globular forms. The occurrence of these forms is quite easily
predicted during the normal growth curve of the organism. (Morton )
There has been, moreover, no correlation between any of the vari-
| ous morphological forms and the virulence of the diphtheric bac-

illus, although the granular type seems to predominate in clini-
cal diphtheria . (Belding and Marston; Morton)

The flagellated bacilli are most often observed without the

TO - j OJ ix r -« ' !

.

i

.

,

1

16

>

flagella. This is the general case, because these structures,
being verv delicate, are often lost due to handling during the
preparation of a stained smear. This, of course, is not a true
variation. However, flagellated bacilli, e.g.. Salmonella para-
tvbhi. mav be caused to show a non-f lagellated variant bv grow-
ing the organism in serum containing antibodies directed agains-
the flagella. This gives rise to a non-f lagellated , non-motile
end non-snecific strain of Salmonella paretvphi . ( Arkwright and
Pitt) This method of producing artificial variations in flag-
ellated or encapsulate L species has been used to a great extent
in the studv of bacterial dissociation, and w ill be mentioned
again later, particularly in reference to the transformation of
pneumococcal tvpes . This variation associates itself with othe:
which are interdependent unon it. Thus, in the experiments de-
scribed bv Arkwright and Pitt, the flagellated form of Salmonei;
paratvphi was determine bv its agglutination in homologous
immune serum containing antibodies directed against the "H" , or
flagellar antigens. Those organisms so agglutinated were ob-
serve to form smooth, dome-shared colonies, show; uniform turbic
itv inbroth culture, and absence of agglutination in 0.8 5% salt
solution. This was designate- 1 as the smooth, or ”S" form,
according to Griffith. Those organisms showing irregularitv of
the surface and margins f the colonies, granular growth in
broth culture, and egglutir abilitv in 0.85 o salt solution were
designated as rough, or "R" forms. Bv the experimental method c
scribed above Arkwright and Pitt cultivated and described variar
forms of o&lmonella paratvphi and m-berthella tvphosa . The smool

s

a

e-

t

h

.

>

»

17

forms of these organisms we re agglutinated, in immune sera conta idl-
ing the homologous "H" antibodies , while the rough forms were
not agglutinate by these sera, but did react when placed in im-
mune sera containing the antibodies directed against the homolo-
gous ''0", or somatic antigens. It was possible to implv from
these observations that the smooth forms of the organisms were
flagellated , while the rough forms had lost their f lamella.

Similar experiments were done with Klebsiella pneumoniae ( Julian-
elle) and various pneumococcus t^ues .( Griff ith ; Riemann) But
whereas it has now be^ n found possible to cause the reversion of
R pneu '.ococci to the original tvne-snec if ic S forms bv growing
the R form in anti-R serum, and bv other methors which will be
described later, it has not been possible in vitro in the case
of the flagellated organisms. The explanation of this phenome-
non is probablv the fact that the R to S dissociation of the
pneir ococcus is due to the elaboration of a polyseccharidal cap-
sule, which appears to be brought about easily by certain trans-
forming and inducing substances present in normal and immune
sera. On the other hand, to convert the R form of flagellated
organisms to the S form it ir- necesserv that a flagellum be
elaborated bv the organism, a change which would appear to in-
volve a fundamental change in the cell structure of the bacteri
urn. This does not appear feasible at the present time.

It has been noted( Julianelle and others) that S forms of
flagellated organisms agglutinate spontaneous lv when treated
with homologous immune sera, whereas the corresponding R forms
agglutinate quite slowlv, giving in higher concentrations a

.

'

.

/

18

>

>

fluffy precipitate, and in the higher dilutions a granular pre-
cipitate, which is difficult to read without a lens.

It very rarely happens that a bacterial cell will exhibit a
single phase of variation. Variations involve changes in the
structure of the bacterial cell. These changes may be visible
or invisible, and further may be permanent or reversible, either
by natural or artificial experimental methods of cultivation.

The change may be merely the loss of a capsule, or it may be a
change of reactivitv towards the gram stain. No matter what it
may be, it usually affects the entire definitiveness of the pro-
perties of the cell. In 1922, De Kruif described an experiment
involving the mutation of the bacillus of rabbit septicemia from
the highly virulent D form to the almost avirulent G form. He
observed, aside from the difference in virulence exhibited by
the two mutant forms, differences in colonial formation and ap-
pearance, and differences in the acid agglutination optima of
the two varieties. He further observed that this last phenomen-
on implied a distinct change in the bacterial protoplasm of the
G form, which would seem to be the most fundamental mutation so
far described. He shov;ed that growing a pure-line strain of the
u organism in diluted rabbit serum enhanced the tendency of mut-
ation to the G form. On the other hand, all attempts to grow a
pure-line strain of the G mutant in undiluted rabbit serum,
which inhibits the tendencv of the D form to change to the G
variety, and thus bring about a reversion to the D form failed,
further confirming the concept of a fundamental change in the
bacterial protoplasm.

.

.

.

19

COLONIAL VARIATION

Colonial variation appears to be the most fundamental mani-
festation of the variability of bacteria in regard to seemingly
permanent changes, because of the fact that it is generally as-
sociated with mutations of almost every variety, one being except-
ed, the artificial transformation of the pneumococcus from one
tvpe to a virulent pneumococcus of another specific type, which
is distinct and transmissible. We shall mention this later.
Colonial variation within a species indicates the occurrence of
mutant forms. Motile varieties may often be distinguished from
non-motile strains, the colonies of motile organisms usually
showing spreading growth, or colonies with irregular margins, as
compared with the usual-# distinct and small colonies of the non-)
jnotile varieties, e.g., Proteus vulgaris .

Non-motile bacteria which tend to multiply in chain forma-
tion may exhibit non-chain forming mutants. The presence of
chains within a colony ma# give a veined or coiled appearance.

This has been shown in photomicrographs (X300 ) by Bisset. Colon-
ies which show such formation are encountered in certain species
of streptococcus; the viridans strain gives a sw/irleci vortical
appearance; Bacillus anthracis. shows the "medusa-head" colony
formation characteristic of that spec ies . (Bisset )

Colonial formation is largely dependent upon the structure
of the organisms of which the colony is composed, and to a lesser
extent upon their biochemical and physiological properties. Col-
onies of encapsulated organisms, as the pneumococci and Klebsi-

,

• t •

.

.

.

20

t

>

ella pneumoniae, are moist and glistening, of stringy consisten-
cy, and possessing even surfaces and margins. It has been
pointed out that the encapsulated S pneumococci should be class-
ified as forming M, or mucoid colonies .( Dawson) This type of
colony is tvoifieci by the encapsulated form of Klebsiella pneumo-
niae, which upon nutrient solid meuia forms rather large, flat,
moist, glistening colonies with even margins. The organisms in
these colonies are rather short forms, even the bacillarv organ-
isms. The consistency of the colony depends upon the amount and
the nature of the caosular substance secreted by the organism.
Klebsiella pneumoniae elaborates a larg amount of capsular sub-
stance, which is easily recognized under the microscope by ord-
inary staining methods, and its colonies are quite viscous. On
the other hand, the pneumococcus elaborates less capsular mater-
ial, hence the mucoid -phase is less viscous. It should be noted
here, and will be referred to extensively later, that the tvpe-
soecific pneumococci which are designated as the S form of the
snecies are actually the M, or mucoid variant, according to thei]
colonial morphology. Colonies of virulent, encansulated pneumo-
cocci are small mucoid varieties. Also, the R form of avirulent
non-ca^sulated nneumococci form typical smooth colonies of the S
variety. It is possible to obtain typical R colonies of pneumo-
cocci (Dawson; Shinn) with a wrinkled appearance, showing fila-
mentous growth. This is found to be the ultimate R cultural
phase of the -pneumococcus .

It has been found possible to relate colonial organization
to the method of growth of the organisms . (Bisset ) It is noted

.

.

.

.

21

that in smooth colonies of motile organisms the colonies are
made up of long individual organisms. This is attributed to the
sliding movement of the organisms ast each other directly aftei
fission. Colonial morphology has be^n correlated with two othei
types of nost-f iss ional movement, snapping and ’ hiding, and the
sliding movement mentioned above. Organisms occurring in the
"medusa-head" colony have been shown to exhibit e snapping nost-
fissional movement. This was first associated with the rough
dissociation chase of Bac illus an three is . (Bissett ) It • as also
shown to occur in the rough colonies of Escherichia coli . Rougi
colonies seem to show a characteristic arrangement of the bactei
ia v hen examine:; microscopical!^, bv reason of the snapping
post-f iss ional movement; the'1’- are generally found lving side by
side in small bundles. The smooth variant forms of coliform
organisms show no che recteristic arrangement within the colony.
Following f is.' ion these organisms separate and slide partially
ast each other. The corvnebacteria show a whipping post-f issic
al motion which gives rise to vortical colonies exhibiting
mrov th in chains similar to colonies of streptococci. It has
been noted (Gause) that Bac illus mvcoides contains an identifiab]
protop' asmic property which causes the threads to grow clockv ise
or counter-clockwise , giving rise to dextral and sinistral
mutations .

It has been observed that practically all bacterial culture
exhibit several types of morphologically distinct colonial forms
including carefullr prepare single-cell cultures. These become
stabilized through several generations and settle down in one

n-

e

s

9

*

.

>

phase or another.

Colonial morphology mav usually be associated with other
structural and biochemical properties of bacteria. These organ-
isms whose virulence depends upon their possession of a specific
capsular substance, such as iilebs iella •pneumoniae and the type-
specific S pneumococci, form k, or mucoid colonies. Hence mucoic.
colonies may be judged as being composed of encapsulated and
virulent organisms. Similarly, when these two species are grown
in immune sera directed against :heir respective specific soluble
substances (Heidelberger and Avery), non-ca sulatea, avirulent
bacterial cells arise which form typical smooth colonies.

In the case of organisms whose virulence is associated with
their flagella, as in the cases of Salmonella paratyphi and 5b-
erthella tmho a , the virulent flagellated motile variants are
shown to form topical smooth colonies ( there are instances of
rough variants). The avirulent, non-f lagellated , non-motile
mutants exhibit rough colonial f ormat ion ( smooth colonies have
been noted). Smooth colonies in general indicate the virulent,
type- or group-specific variation of the species; rough colonies
are almost invariably avirulent and species-s >ecific . (Arkwright ;
Bruner and Edwards; Morgan and Beckwith)

The diphtheria organism shows a variation in its colonial
organization which appears to be correlated with varying degrees
of virulence . (Morton) Three strains of the virulent form of
dor^nebacterium diphtheriae are recognized by colonial formation
on blood-tellurite medium. These are designated, according to
the original theory of their varying degrees of virulence, mitis ,

22

. ■

.

intermedins , and gravis . However, it has been found that these

original designations do not truly connote the relative virulenc
of the three strains . The morphologv of the gravis strain con-
sists of uniformly staining short forms, which have been found
in some clinical cases, especially in the more severe ones.

Its exact role in clinical diphtheria is still not yet clear.

The gravis strain shows two main antigenic groups, each of which
ha- been found as an eoidemic strain over wide areas. The
mor diology of the mitis strain sho\ s long slender forms with met|i
chromatic granules. This type is the predominant one in clinica
diphtheria in this country. It exhibits a great aeal of anti-
genic diversity, a fact 1 hich nay account for its greater over-
all virulence and occurrence . (Belding and Marston; McLeod) The
intermedius type exhibits the barred clubbed variety of the

bacillus, and it is an anti enically homogeneous group. The
three strains also exhibit some variation in their biochemical
properties, in that the gravis strain will ferment starch and
glycogen, while the mitis and intermedius will not.

The snecies Klebsiella nneumor iae v as observed (Julianelle)
to be divided into three distinct serological types, A, B, and C
These three types, in their virulent, encapsulated phase, are
agglutinated only in their homologous antisera. The virulent
encapsulated variants of each t^oe form M, or mucoid colonies on
nutrient agar elates, k.hen these colonies are transplanted for
several weeks, a. translucent mutant arises in each case. The
number of transplants necessary to bring about the degradation
varies with each type. Type A needs about four weekly transplants

23

.

.

24

>

while Tyne B needs six to eight weeks, and Type C will usually
show the mutant form after only two weekly transplants. If the
procedure of subculturing is continued, the smooth, translucent
variants give rise to a still further dissociated phase, the
rough variant. The translucent colony variant is smooth, and
composed of short non-ca psulated rods, resembling those of the
parent strain in each type. The translucent variant has always
been the first to appear during the process of natural dissocia-
tion. The rough variant is invariably derived from the trans-
lucent type and produces a rough appearing colony which is com-
posed of long non-capsulated rods and filamentous forms.

The M, or mucoid phase of Tvoe A is designated as As, the
smooth phase as At, and the rough or R form as Ar, those of Type
B as Bs, Bt, and Br, and of Type C, Cs, Ct, and Cr. Agglutina-
tion te ts were performed using antisera, preparec. against each
of the nine mutants of Klebsiella pneumoniae. It 1 as shown that
all the mucoid forms w'ere agglutinated only by their respective
homologous immune sera. In each type the smooth variants v/ere
agglutinated by their respective immune sera, and each showrea
cross -agglutination with the immune serum prepared against the
mucoid phase of the homologous ty e. The rough variants Ar and
Br were agglutinated only by anti-Ar and anti-Br sera, respect-
ively. The Cr mutant was agglutinated by its homologous immune
serum, and showed cross -agglutination with anti-Ct serum. Juli-
snelle (1937) has reported that the rough variants of Klebsiella
pneumoniae of different serological types cross-react. The resu
of investigations by Randall (1939) do not confirm the occurrenc

Lts

3

.

4

.

of cross -reactions among the rough variants of encapsulated
Klebsiella pneumoniae. It is probable that the use of two dif-
ferent methods of inducing dissociation explains the different
results, and as shown by Julianelle even rough variants derived
from the same serological type 02 Klebsiella pneumoniae are not
always serologically identical.

Hence we mav now describe three distinct tvnes of bacterial
colonies, each of v/hich appears to be linkea with one or more
variations of the structural, and possible of the biochemical
properties of the organism.

The M, or mucoid colony consists of encapsulate a , generally vir-
ulent organisms, giving specific agglutination in homologous
immune serum, ana no cross-agglutination with immune sera of any
other t^pe within the species. The colony is viscous, moist,
glistening, and possesses even margins; without animal passage
it has a tendencv to dissociate towards a more "stable" form.

In this class are found the S forms of the pneumococcus.

The S, or smooth colony consists of non-capsulated , gener-
ally avirulent organisms. This type is found only in those
soecies v/hich exhibit a mucoid phase, -which is true of most
pathogenic bacteria. The other group of organisms forming S
colonies are generally flagellated organisms, the Shigella and
the Coccaceae being the principal exceptions. These forms, like
the mucoid variants, tend to dissociate to more "stable" forms.
The S phase forms smooth, generally moist colonies, with even
surfaces and margins. The R form of the pneumococcus falls unde

0

>

26

this classification of colonial variation.

The Rr or rough colony is made up of non-capsulated , or nonf-
flagellated, avirulent organisms (except Bacillus anthracis,
which is virulent in- the rough colony phase). The organisms in
this ohase are species -specif ic only (exceptions are noted in
the hnterobacteriaceae ) . The ultimate R cultural Phase of the
pneumococcus is the true R form of the species.

It is difficult to draw up such general definitions for eac
of these dissociative phases, for there are within a species or
a species type, and from one species to another, variations
which lie outside these general rules. There are the encapsul-
ated S and R variants of Klebsiella Pneumoniae ( Julianelle ) , the
smooth non-motile ana rough motile variants of the Colon-Salmon-
ella organisms (Arkwright; Kinsser; and others), and the virulen
R forms of the pneumococcus . (Griff ith) These, along with all th
other naturally occurring and artificiallv induced, and trans-
formed mutations, go to make up a very heterogeneous group of
organisms indeed. There are so many factors of environment,
both purposed and accidental, that the subject must more or less
be studied not only from the stanapoint of individual species
alone, but of single variations within species, their causes andj
the-ir effects, ana the factors which may be introduced into any
system to cause reversions or further dissociations, or perhaps
still more fundamental bacterial changes.

i

27

VARIATIONS OF TEN PNEUMOCOCCUS

The study of the oneumococcus in its many nhases of dissoc-
iation has made up a large part of the literature on the oheno-
mena and mechanisms of bacterial variation. The exneriments set
forth by the various workers (Allow ay; Avery; Brown; Dawson;
Dochez; Dubos ; Eaton; Goebel; Griffith; Heidelberger ; McCarty;
Mudd; Paul; Riemann; Shaw; Shinn; Sia; Strvker) represent studies
of both the naturally occurring and the artificially induced
mutants. The variations to be discussed are those of colonial
formation, innuno chemical specificitv, and the transformation of
pneumococcal tvnes ; also, some of the methods which have been
used in the course of the. e investigations an 1 1 ich are annlie-
able to other biological problems will be discussed.

The wor on the nneumococci as a groun, subsequent to the
discover^ and cultural isolation of the organism, bTr rasteur in
1881, and by Fraenkel in 1886, resnectivelv, was begun in 1905
by Collins, hiss, and Park and Williams, who did extensive
studies of the biochemical Droperties of the organism, and exa-
mined the behavior of the various strains with which they worker
towa rd immune sera. The work received its greatest impetus
from the observations of Neufeld and Levinthal vdio in 1909 sep-
arated all of the then-known strains of pneumococci into four
distinct serological groups, Types I, II, and 111, and the heter-
ogeneous Groun IV, by means of the capsular swelling! quellung )
reaction. They observed that when pneumococci were mixed with
immune sera prepared bv injecting rabbits with living inocula o'

28

various strains of pneumococci, the capsules of the organisms
became swollen in certain sera, those which had been prepared
against the homologous organism. They were able to show that
those organisms which showed capsular swelling in one immune
serum showed it only in that serum, i.e., the reaction was tvpe-
soecific. This method was disregarded for a number of vears due
to the inability of subsequent workers to duplicate the original
results. Recently (1943)> Mudd, Heinmetz, and Anderson have
been able to repeat the original experiments, using immune sera
prepared from rabbits, and have prepared a series of electron
photomicrographs of the reaction in an attempt to explain more
fullv its mechanism. At the present time this method is widely
used, due to improved methods of ore caring the specific immune
sera from the blood of rabbits, and it has replaced the formerly
used methods of Krumwiede and Sabin. The original Types I, II,
and III are still so designated, while the heterogeneous Group
IV has been separated into over seventy ty es , by means of ab-
sorption techniques. All the types so examined and classified
were gram positive cocci, occurring in pairs or in chains, and
exhibiting identical structural properties, i.e., the possession
of a capsule, virulence to mice, solubility in bile salts and
similar chemical compounds, inulin fermentation, and the produc-
tion of methemoglobin from oxyhemoglobin.

In his studies on patients suffering from pneumonia, Grif-
fith (1928) described two colonial variants of pneumococci iso-
lated from both the sputum specimens of hi s patients and the
pleural exudates from experimental mice which he was using for

I

(

.

.

<

typing. The one was the virulent, encapsulated organism which
formed moist, glistening mucoid colonies. These he designated
as the 3 form of the pneumococcus , adopting Arkwright's termin-
ology. He also described a rough variant which was avirulent
and non-caosulated . These pneumococci formed smooth, moist
colonies on blood agar plates, and were designated by Griffith
as the R form of the pneumococcus. This nomenclature has led to
a great o.eal of misconception on the part of workers who are not
familiar with the organisms involved.

It has been demonstrated (Avery; Dawson; Griffith; Reimann;
and others) that the S form of colony is invariably composed of
virulent type-specific organisms, and that this property, i.e.,
type-suecif icity , is transmissible indefinitely through mouse
passages and subcultures on normal serum media. There is a
tendencv on the part of these S organisms to dissociate towards
the intermediate SR and Rs forms, and towards the ultimate R
cultural nhase (Shinn) \ hen grown on non-vital media. However,
several animal passages of anv organisms showing this tendency
will usually suffice to restore their virulence and type-speci-
ficity, and the formation of typical mucoid colonies. In con-
trast, the R pneumococci show species-specificity only, i.e.,
the^ are agglutinated by the same sera regardless of the S type
from which they were derived, and transmit this property in
series through innu lerable transplants. However, if such forms
are passed through mice, there may be a reversion to the 3 form
from which thev were derived. This will depend upon the degree
of dissociation. It was shown (Alloway; Dawson and Sia) that R

.

.

4

30

pneumococci which were grown in media containing serum against
the species antigen of the pneumococcus showed a tendency to re-
vert to the original virulent, encapsulated S form from which
they were derived.

All S pneumococci are virulent when injected into mice, and
the organisms which are isolated from the pleural exudates upon
autonsy (Heidelberger ) are shown to he serologically and coloni-
ally identical with thoses of the original inoculum. Conversely
all R pneumococci are non-oathogenic to mice. An exception to
this rule was noted bv Griffith, who reported finding only R
pneumococci in the pleural exudates of a mouse which had died
after having received an injection of living R pneumococci togeth-
er with a heat-killed vaccine of Tyne III S pneumococci. No ex-
planation has been given.

All type-specific, virulent ana encapsulated S pneumococci
which form typical mucoid colonies on blood agar plates, and
will retain the aforementioned properties through an indefinite
number of mouse passages. However, when these organisms are
subjected to a lengthy series of transplants and subcultures in
broth media anu on blood agar plates, they tend to dissociate
towards the R form, thus losing their virulence for mice, their
type-specificity, and their ability to elaborate the capsular
substance, depending upon the degree of dissociation, these
properties may be recovered by various means, e.g., growth in
anti-R serum. The R form, on the other hand, when grown on normj-
al serum-containing media, retains the R phase, and show no
tendencv to revert to the S form. There may however be a furthe

.

t • •

31

dissociation tov/ards the ultimate R cultural Phase.

In 1933 Dawson described a rough variant colony of the
pneumococcus which exhibited marked pleomorohisin. Many elongateld
coccoid and coccobacillarv forms were seen. Organisms from earlfy
cultures were entirely gram positive, but smears taken from
older transplants showed a marked variation in reactivity
tov/ards the gram stain. He notea that in a long series of traps
plants these pleomorphic cells had a tendencv to revert to the
rough variant of the R form and also ultimately to the parent
strain of the S form.

The rough variant of Dawson is coloniallv and morphologic -
ally identical with the ultimate R cultural phase of the pneumo-
coccus, which was describee bv Shinn in 1937* Hov/ever, Shinn
reported that there was no tendencv tov/ards reversion of this
ultimate R form, i.e., highlv dissociated, to typical R or S
pneumococci. He also reported that this highlv dissociated var-
iant was no longer soluble in bile. These differences would
seem to indicate that the ultimate R cultural phase of Shinn
was further removed both morphologic allv and phv: iologicallv, arj|d
in regard to colonial formation from the virulent, encapsulated
mucoid form of the pneumococcus than the variant described bv
Dawson. It appears that the ultimate R cultural phase is the
true rough variant of the species, since it has reached a high
state of stabilitv, and cannot be reverted bv any means to
either the variant R form or to the mucoid type-specific S form;
furthermore it seems to have undergone a fundamental change in
its protoplasmic and cell-wall structures, being no longer sub-

,

.

. . . . -

■

32

Jject to autolysis (Eaton; Shinn), and showing marked variability
In reactivity to the gram stain. (Dawson) These last facts, and
its loss of the power to elaborate a capsular substance, as well
as the power to put into operation an enzyme or other phvsiolog-
Lcal system capable of producing such a substance, even when a
transforming material is added to the system, point to a funda-
nental change in the biochemical and physiological processes
vithin the pneumococcal cell.

•dawson ’ s further experiments on the colonial variations of
the pneumococcus led to his study of the relationships among the
solonial variations of Streptococcus heraolyticus , an organism
related to the pneumococcus . In this instance, as in the former,
it y as found that three types of colonies could be isolated from
9 lengthy series of transplants, beginning with virulent organ-
isms forming typical mucoid colonies on blood agar plates. These
were quickly dissociated to the smooth, non-capsulated form,
since hemolytic streptococci have been shown to elaborate a cap-
sular substance only for a very few hours after seeding and
streaking a known virulent strain on artificial media. Further
transplants degraded the smooth hase still further and the R
phase was obtained. These three colonial types corresponded
relatively in virulence, encapsulation, and the colonial morpho-
logy to the three types of colonies of the pneumococcal mutations!.
These variations were further correlated to the colonial mutants
of a wide range of bacterial species, Salmonella Paratyphi (Ark-
wright and Pitt), Klebsiella pneumoniae ( Julianelle ) , Bacillus
enthracis (Bisset), Streptococ ms viridans (Bisset), etc.,

.

_

,

In summation of the colonial variants of Diplococcus pneum-
oniae , the M, or mucoid phase of virulent, encapsulated, type-
specific, gram positive lanceolated diulococci, which form moist
glistening, stringy colonies w ith even surfaces and margins.

The S, or smooth phase consists of usually avirulent, non-cap-
sulated, group-specific gram positive diplococci which form
smooth, moist colonies with even surfaces and margins, generally
a little smaller than the colonies of the mucoid Phase. These
pneumococci retain the property to revert to the mucoid Phase
under the proper conditions, or the R form under further condi-
tions conducive to such further dissociation.

The R, or rough phase of the pneumococcus consists of avir-
ulent, non-capsulated , species-specific organisms which show
great variation in their reactivity to the gram stain, sometimes
gram positive, sometimes gram negative, and exhibit marked pleo-
morphism, ranging from the typical lanceolated cocci occurring
in pairs to long filamentous forms. These pneumococci form
rough, drv colonies on blood agar plates, with rough surfaces
and irregular margins. The R form of the pneumococcus is bile
insoluble, and has lost the property of reverting, or being in-
duced to revert to the S or M phases by any means.

34

TRANSFORMATION OF PNEUMOCOCCAL TYPES

In 1909 Neufeld and Levinthal were able to divide the species
Giplococcus pneumoniae into three distinct types and one large
heterogeneous group, on the basis of serological procedures.

The()r designated these subdivisions as Types I, II, and III, and
Grout) IV. Types I, II, and III reacted only with their homolog-

N

ous antisera, though Types I and III did show cross-agglutina-
tion in dilutions up to 1 to 40 0. Group IV was a large hetero-
geneous mixture of types which at that time these workers were
unable to classify on any serological basis.

In 1915 Avery studied a number of pneumococcal strains
which showed agglutination in Tyne II anti -pneumococcal serum.

He showed that these strains, which he subdivided into three
subgroups and classified under Type II, were serologically
distinct from Type II pneumococcus. Type II anti -pneumococcal
serum after having been absorbed with Type II pneumococci was
unable to agglutinate any of the three subgroups. When the same
serum was absorbed with any of the pneumococci of the three sub-
grouns , it was still capable of agglutinating either of the
other two subgroups, and also Type II pneumococci . It appears
here that these strains, which Avery was able to classify in
three distinct serological groups, were part of the heterogene-
ous Group IV. These had previously remained unclassified be-
cause of their cross-agglutinative reactions with Types I, II,
and III antisera. The agglutination, absorption and protection
techniques used by Avery hsve now been applied to a great number

-

.

.

.

.

35

of pneumococcal strains vhieh v/ere lumped, together under Group
IV; this group has been shown to contain over seventy serologic-
ally distinct types .

In 1916 further mutations of Type II pneumococcus were
noted by Stryker. Mutants were brought about by growth in homo-
logous immune serum, and variations in virulence, bile solubil-
ity, inulin fermentation, capsule elaboration, and antigenic
properties were observed. It is worthy of note that of all the
variations that have been observed in association with the
pneumococcus from time to time, the only one which has been seen
b'r direct observation of the organism has been the presence or
absence of the cansule. The importance of the capsular sub-
stance will be demonstrated shortly.

In 1917 Dochez and Avery demonstrated in the cell-free fil-
trates of pneumococcal cultures a specific soluble substar ce
which reacted with anti-pneumococcal sera. They showed that
this substance of itself v.-as non-antigenic . It was further
shown that this substance differed with each S type of pneumo-
coccus, thus correlating it to the serological specificity of
the several types. It was shown that each of these soecific
soluble substances could be isolated from the cell-free filtrate
of pneumococcal cultures, and also from the bacteria themselves
after they had been freed of all culture medium material, fol-

I

lowed bT7- pneumococcal autolysis by various means . (Keidelberger
and Averv; Goebel; Brown)

It is now generally believed that the specific soluble subp
stance of the pneumococcus which is found in the cell-free fil-

.

.

) .

trates of cultures, in urine, serum, and ascitic fluids of in-
fected individuals is identical with the capsular substance of
each type of pneumococcus. On this basis Heidelberger (1923,
1924, 1925, 1927, 1936) isolated the specific soluble substances
of pneumopoccus Types I, II, and III from cell-free filtrates.

He showe these three substances to be similar in chemical com-
position, all of them being essentially polysaccharides of the
order of starch and glycogen, though giving the reactions of
neither of the. e two compounds when treated with iodine reagent.
Thev were shown to yield glucose on hydrolysis, along with other
carbohydrate compounds, among which were an aldobionic acid, in
the case of Type III, and an acetvlated amino sugar, in the case
of Type I. It appeared that each of these compounds, including
glucose, which probably enters in as a factor due to its type of
linkage with the other substances, were at least a part of the
determinative factors which gave to each pneumococcal type its
particular nature and antigenic properties. These substances,
as obtained, were subjected to a variety of qualitative and
quantitative chemical and physical tests. They were shov/n to

f

differ from one another in content of carbon, hydrogen, oxygen,
and nitrogen, and in the optical activities of their aqueous
solutions. The^ appear to be protein-free substances, giving
none of the usual qualitative tests for proteins and mino acids.
Their serological specificity has been shown by means of the
precipitin reaction with immune sera. At the present time some
seventy-five or more distinct serological types of the pneumo-
coccus have been demonstrated by various techniques. Each of

■

37

these specific soluble substances has been isolateu and subject-
ed to chemical and physical analyses . (Blake ; Brown)

The capsular polysaccharide of the pneumococcus has been
shown to be responsible for many of the properties attributed to
the organism; thus variations in the capsule cause correlated
variations in other properties. While it has been demonstrated
that the virulence of the pneumococcus is not attributable to
either an exotoxin or endotoxin, it is readily shown that the
organism is extremely invasive. This high degree of invasive-
ness is largely due to a high resistance to phagocytosis within
the animal bod^. The resistance of the organism to phagocytosis
is attributed to the presence of the capsule, which is a highly
polymerized and very viscous polysaccharide. It is difficult to
say, though it is generally assumed to be so, that the polysacch
aride elaborated by the organism is a true capsule, since the
substance is also found in cell-free filtrates, being water-sol-
uble. It majf be that this substance merely collects around the
organism following its elaboration and secretion through the
cell wall and clings to the wall by reason of its own high vis-
cosity and its internal cohesion, hncapsulateo. pneumococci when
freed of any culture media may then be v ashed free of the cap-
sular polysaccharide, similar to Klebsiella nneumoniae , and in
contrast to Mycobacterium tuberculos is , the waxy capsule of
which appears to be an integral part of that organism, and can
onl*r be removed by such drastic procedures as treatment with
boiling alcohol. On the other hand, serological evidence points
to the fact that the capsular polysaccharide alone is incapable

38

of stimulating the production of antibodies, when it is freed of
the rest of the intact pneumococcus. It has also been shown
that the injection of lysed cells which still hold the protein
substance of the pneumococcus in combination with the polysacch-
■ ride ere also incapable of stimulating antibody roduction.

The cell protein alone will not evoke antibodies against type-
specific neumococci , or against the cell ' eotein combined with
the polysaccharide ; further, the cell protein will not react
with any tvpe-specif ic anti -pneumococcal serum. Or: the other
hand, the polysaccharide alone, or in combination with the cell
protein will give a precipitin reaction with anti: era produced
by the intact cells.

Thus v/e have seen that the capsular olysacrharide of the
pneumococcus is responsible for the resistance of the organism
to phagocvtosis within the animal body, for the virulence of the
organism by way of enabling it further to invade the host, and
for the specificity of the seventv-five serological types. The
other variations, bile solubility and inulin fermentation, 1 while
they ap >ear to be associated with variations in capsular elabor-
ation, do not seem to bo directly dependent upon it, as there
seems to be a change within the bacterial protoplasm which may
ac court for such mutations.

The last variation associat&dnwith the inhibition of cap-
sular elaboration to be discussed is that of colonial formation
of the pneumococcus. Although this sub ect has been taken up
previouslv in some detail, it will be necessary to repeat certain
salient facts pertinent to a further explanation of these muta-

r

.

*

.

39

tions . The pneumococcus dissociates into three distinct chases,
mucoid, smooth, ana rough forms. The colonial distinction of
these three forms may be made macroscopically by observation of
the texture and share cf the colonies, when observed microscop-
ically the organisms aupear encapsulated in the mucoid phase,
and rarely in the smooth ohase; the non-capsulated organisms are
found in the smooth and rough phases. The smooth and rough
phases may be differentiated from each other microscopically by
showing that the smooth phase consists of regular morphological
forms, i.e., dirlococci, v.hile the ultimate rough phase consists
of long rods and filamentous forms, which often appear gram neg-
ative. The further serological distinction was made that the
smooth rhase could be reverted to the mucoid phase, while the
rough phase no longer retained this prooerty.

We have also seen previously that when the smooth phase, or
so-called R form of the oneumococcus was reverted to the mucoid
phase, it again required the serological snecificity of the S
tvpe from which it originally dissociated . (Dawson; Stryker)
However, it has been found oossible, by means of the intermediate
smooth phase, or R form, to convert pneumococci of one tyre to
those of another type, and this specificity may then be trans-
mitted in series, or ma^ again be reverted to the original type,
or to still another specific type, again by way of the inter-
mediate smooth ohase.

In 1923 Griffith; described the effect of culturing tvpe-
soecific nneumococci in homologous immune sera. From these cul-
tures he obtained the R forms of the respective soecific types.

.

.

'

.

j

.

40

>

These rough phases were identical with respect to their biologic
al properties and serological manifestations, except for the fac
that unon reversion to their original mucoid chases they regaine
only the type-specificity of the parent strain from which they
had been derived. The work on the transmutation of pneumococcal
types has stemmed mainly from the observations of Griffith (1928)
and Riemann (1929). In an epidemiological study undertaken in
1928 on a number of patients suffering from pneumonia, Griffith
studied the pneumococcal types in sputum specimens collected
from these patients. Among them he found pneumococci of Types
I, II, and III, of Group IV, and a species-specific type which
has been described previously, and which was designated by that
investigator as the R form of the pneumococci . After this pre-
liminary scutum typing, Griffith Proceeded to pass the organisms
through mice, and then to type the organisms obtained in the
peritoneal washings of the animals at autopsy. He found that
when certain sputa which had been identified as containing Type
II organisms ere injected into mice, the animals apeeared to
have died from an infection of Type III 3 pneumococci, which
were obtained from the peritoneal v,ashings in almost pure cultur
This apparent change in type was flso associated v;ith a
definite variation in the viru ence of he organisms. This was
additional evidence, showing that the Type II 3 pneumococci had
acquired not only the serological soecificitv of Type III S
organisms, but had acquired the virulence of Type III. The
change in virulence is undoubtealy due to two properties of the
pneumococcus, first, that the Type III pneumococcus is a more

.

.

.

.

41

virulent organism than Type II, and. second., upon which the first
may very well depend, that Type III pneumococcus elaborates more
capsular polysaccharide than Type II, and in its virulent phase
it tends to forms chains of encapsulated diolococci, thus com-
bining the invasive -cowers of both the streptococcus and the
pneumococcus with an increasec resistance to phagocytosis .

Riemann had found previouslv (1925) that type-specific S
pneumococci grown jin vitro in immune serum of the homologous S
type gave rise to a mutant form consisting of non-type-specific
non-capsulated R forms of the pneumococcus , which retained those
properties upon subsequent transplantation and subculture in
either the homologous S ty:>e immune serum or in normal serum.

He further noted that when these non-specific R pneumococci were
injected into a mouse together with a heat-killed suspension of
S pneumococci of the homologous type from which the R mutant had
been derived, the mouse died of an infection with living t^pe-
specific 8 pneumococci. Upon serological examination th8ee 8
pneumococci were identified as being of the same tvpe as the
organisms in the heat-killed suspension. The organisms in the
shepension were shown to be non-viable, and the living R pneumo-
cocci were demonstrated to be of themselves avirulent to mice,
by suitable animal and culture medium controls. It has been
shown that the virulence of the pneumococcus is to a large exten
dependent upon the individual ca sular substance. Conversely,
the R pneumococci have been shown to be non-capsulated and avir-
ulent. The former property appeared to be dependent upon the
latter. It has also been noted th t the R pneumococci show no

b

'

.

42

agglutination or swelling reactions in any of the type-specific
sera. It was subsequently shown that all R pneumococci, regard-
less of the S strain from which they were derived, are agglutin-
ated only in anti-R serum, regardless also of the derivation of
the R strain used to prepare that serum.

From the above data the assumption was made that there must
be in the heat-killed suspension of 3 pneumococci a substance
which is capable of inducing the elaboration of capsular poly-
saccharide by living R pneumococci, which of themselves did not
possess this irooerty. It was further assumed that this induc-
ing substance was type-specific, i.e., it induced the elabora-
tion of the capsular polysaccharide of the specific type from
which the transforming substance was derived, and of that type
only. This was sho’ m in further exoeriments (Avery; Dawson;
Dawson and Sia; Reidelberger; McCarty; McLeod and McCarty) in
which R pneumococci derived from a number of different S types,
I, II, III, IV, and VIII, we re reverted to their original type-
specific it r, and furthermore, could be converted to any of the
other tvpes, by the use of cell-free filtrates of the desired
type, and by the use of the intermediate smooth nhase. The pro-
cedure used was that originally propounded by Griffith. Dawson
injected living R pneumococci derived from Type II pneumococci
into mice, together with large amounts of a heat-killed vaccine
of Tyne III S pneumococci. When these organisms were isolated
from the pleural exudates (Heidelberger found this to be a more
convenient and more abundant source of the organisms than the
peritoneal washings used previously) subsequent to the death of

.

.

the animals, they were shown to agglutinate in Type III anti-
serum, and to give pure cultures of Tyne III S pneumococci when
cultured in serum media . Similarly, when a heat-killed vaccine
of Type I S pneumococci was substituted for the Type III S sus-
pension in the above experiment, the organisms isolated were
shown to be Type I S pneumococci. In neither case were any S
pneumococci of Type II isolated from the animal exudates.

It is therefore seen that S pneumococci of Type II have
actually been transformed, by means of a specific transforming
substance, into S pneumococci of Type I, and of Tvne III. In
each case the transformation v as brought about through the use
of the interne date smooth phase of the organism. A number of
workers (Alloway; Dawson ; Dawson and Sia), by the use of cell-
free filtered extracts of Type III a pneumococci, succeeded in
effecting the transformation of Tyne II S pneumococci into Tyre
III S pneumococci in vivo and _in vitro ; the _in vitro transform-
ation was brought about through the use of the intermediate R
form, as were the transformations in vivo. There is non sub-
stantiated data showing the direct conversion of S pneumococci
of one type to S nneumococci of another snecific type. Barnes
and Wight have described the snontaneous transformation of Type
V S nneumococci to Tyne II S nneumococci _in vivo , but this work
has never been corroborated .

By the injection of a small inoculum of living R forms of
Type II nneumococcus , together with large amounts of a filtered
cell-free extract of Type III S pneumococci into a mouse, it has
been possible to obtain a pure culture of Type III S pneumococci

.X

Uk

Similarly, Type II S pneumococci were converted to the R phase
by growing the organisms in Type II anti- oneumococ cal serum.

The organisms thus obtained were then transplanted to a culture
medium containing a filtered cell-free extract of TYPE III S
pneumococci. Similar transplants were made of the Type II R
organisms to media containing filtered cell-free extracts of
other type-specific S pneumococci. In each case the Type II R
pneumoc ;cci were converted to the S form of the homologous sero-
logical type of the filtered extract used in the culture medium.

These experiments led various investigators (Avery; Keidel-
berger; LQLeod; McCarty) to the conclusion that there must be
present in the filtered cell-free extracts of S pneumococci a
substance, which has been produced by the specific type of org-
anism present in such an extract, and which is capable of induc-
ing the elaboration of the homologous type-specific capsular
poljrsaccharide bv R pneumococci, regardless of the type deriva-
tion of the latter. It may further be concluded that R pneumo-
cocci, while they do not possess any of the several transforming
substances in situ, nevertheless retain the property of utiliz-
ing such a substance, when it may be added to the system. This
may at some future time be shown to be due to the activation of
an enzyme, or of an enzyme system, which is responsible for the
elaboration of the capsular polysaccharide.

There is, however, the aforementioned ultimate R cultural
phase of the pneumoco cus, described by Shinn, and by Dawson.

It was found in many instances that when S pneumococci were
transformed, or "degraded”, to the R form, and were subsequently

.

.

45

injected into m?.ce, together with heat-killed vaccines of either
homologous or heterologous 8 pneumococci, there appeared to be
no reversion or transformation of R to S pneumococci. An in-
crease in the amount of extr-ct or vaccine used for injection
seemed to have no effect upon the reversion or transformation.
Shinn continued his experiments by growing the organisms in
homologous anti-S serum, thus transforming S to R pneumococci.

He continued this degradation through a long series of trans-
plants in this anti-S serum. He finally reached a point of
dissociation at which he had obtained a pure culture of aviru-
lent and non-capsulated R pneumococci, which could not be trans-
formed t any other type-soecif ic S pneumococci, nor could it
be reverted to its original ty ;e-s pecif icitv by any means. No
capsule formation could be noted, or induced by any means, nor
was there any virulence for laboratory animals. This was termed
the ultimate R cultural chase of the nneunococcus , and as such
was discussed in detail in the section dealing with the colonial
variations of the pneumococcus.

.i

• -

. Hr*

4 6

NUCLEOPROTEINS AND BACTEuIAL GENETICS

It was not until 1944 that a more extensive study was made
by Avery, McLeod, and McCarty of the transforming substance that
was known to be present in pneumococcal cultural extracts. The
work of the above investigators was directed mainly towards the
isolation, purification, and examination of the chemical and
physical properties of the transforming material, as well as its
physiological behavior. They started with crude extracts of
cultures of Pneumococcus Type III. (This work has since been
extended to two other pneumococcal types, namely, Types II and
VI.) These crude extracts from heat-killed pneumococci are com-
plex mixtures of components of the bacterial cells themselves,
along with the substances originally present in the culture
medium. It was found that the removal of proteins, lipids, the
somatic and capsular polysaccharides, and the ribonucleic acid
fraction from extracts of Type III cultures had no effect upon
the transforming activitv of the residual extract material.
Accordingly, they -proceeded to remove these components as quant-l
itatively as possible, the proteins by the chloroform method,
the somatic carbohvdrate by fractional alcohol precipitation,
the capsular polysaccharide by digestion with a specific enzyme
which hydrolyzes it, and the ribonucleic acid bv digestion with
the enzyme ribonuclec se , or by alcohol fractionation. (Levene anc
Bass )

The product obtained after these preliminary extractions
Lad been carried out possesses practically all of the biological

'

47

activity oi the original crude extract. It was readily soluble
in aqueous and saline solutions, giving clear, colorless solu-
tions which were highly viscous even at relatively low concentra-
tions, and which showed high bifringence. The material is pre-
cipitated by alcohol in the form of a mass of fibrous threads,
which loses none of its tran forming activity upon repeated al-
cohol precipitation. Qualitative tests for both proteins and
ribonucleic acid are negative. On the other hand, the diphenyl-
amine reaction for desoxyribonucleic acid is strongly positive.
The quantitative elementary tests carried out on several differ-
ent samples of the active material sho the content of carbon,
hydrogen, oxygen, nitrogen, and phosphorus to be comparable to
that found in pure samples of sodium desoxyribonucleate . The
nitrogen-phosphorus ratio of 1.6? conforms closely to that of
the sodium salt, and further eliminates the possibilitv of pro-
tein contamination. Absorption spectra curves in the ultravi-
olet region showed a maximum high absorption at 2600 A, which is
known to be characteristic of nucleic acids . (Caspersson; Mirsky)
Solutions of the material lose none of their biological activity
after treatment with the enzyme ribonuclease . These data lead
therefore to the conclusion that the transforming activity of
the material is associated with a nucleic acid of the desoxy-
ribose type.

Solutions of the purified substance were not observed to
give precipitin reactions with Tyne III anti -pneumococcal serum
of high titer, in the dilutions characteristic of the serologic-
ally active substances, though slight reactions we re noted in

J-- . cr

,

t ib ■ ov; >. - v. <

.

.

.

48

>

very low dilutions.

Analysis of samples of the purified material in the Tiseliu
apparatus and in the Svedberg ultracentrifuge both give evidence
uointing to the nresence of a homogeneous substance of high
molecular v right. The ultracentrifuge gave a sedimentation
constant corresponding to a molecular wieght of apnroximately
1,000,000, which is characteristic of nucleic acids of the
desoxvribose type. The Tiselius electro- horetic data also
showed a single sharp boundary, with which the transforming
sctivitv appeared to be associated.

The transforming activity of the material was tested quant-
itatively with serial dilutions of R pneumococci derived from
Tvpe II S organisms. These were in each case transformed to
Type III S neumococci. It was found that the substance was
active in a final dilution of 1 in 6C ,000,000, c.2 milliliters
being used, which contained 0.003 microgram the unified

substance.

It v/ as thought that oossibly the substance which w as actu-
ally responsible for the activity of the transforming substance
was still unknown, and that it might be adsorbed, or in some
other manner be attached to the desoxyribonucleic acid molecule.
Tbe investigators therefore looked about for a suitable specific
biological tool with which they might destroy or inactivate the
desoxyribonucleic acid fraction of the transforming mixture, if
indeed there be such a mixture, thereby leaving the true trans-
forming subs tar ce intact. It had long been known that enzymes
were among the most specific of biochemical reagents at our

3

.

.

.

.

49

disposal at present. These organic catalytic agents are even
more specific than antigen-antibody reactions. It had been
noted, for example, that Type VIII S pneumococci showed cross-
reactions in the presence of Tyne III S anti -pneumococcal serum. ,
During the study of this phenomenon, Dubos had isolated from a
bacillus found in the soil, the S III bacillus, an enzyme which
was capable of destroying the capsular polysaccharide of pneumo-
coccus Type HI, and which had no effect upon the capsular poly-
saccharide of Type VIII. Similarly, an enzyme was found which
attacked only the Type VIII specific polysaccharide, leaving the
Type III specific substance intact . (Sickle and Shaw)

In 1940 Kunitz described the method of isolation and the
properties of an enzyme, ribonuc lease , which he had found in
beef Pancreas. This enzyme was found to be specific in its
action upon ribonucleic acid (yeast tv e), having no effect upon
desoxyribonucleic acid (thvmus type). In 1944, McCarty, also
v orking with beef pancreas, isolated an enzyme which was found
to be specific in its action upon desoxvribonucleic acid, and
was ineffective against nucleic acids of the yeast type. This
enzyme was called desoxyribonuclease . It was this substance
that was used to determine whether or not the desoxyribnucleic
acid fraction isolated from Type III S pneumococcal cultures
was actually the inducing substance of the transforming material,
or whether it acted me ely as a carrier for that substance.

From the previous experiments, particularly the electrophoretic
and ultracentrifugal methods, it was established that the active
substance was a fairly homogeneous material. Accordingly, ex-

.

*

.

.

50

>

periments were set up to determine whether or not the fraction
isolated from the pneumococcus containing the activating prin-
cinle was indeed inactivated after treatment with the new enzyme
desoxvribonuclease . It was found that as little as 0.01 micro-
gram of the enzyme completely inactivated the transforming sub-
stance. This evidence led quite conclusively to the fact that
the actual transforming substance involved in the transformation
of pneumococcal types was a nucleic acid of the desoxvribose tvpe,
and, moreover, that the substance isolated from the pneumococcus
cuitute extracts was a highly purified com ound.

The inactivation of desoxyribonucleic acid mav be brought
about by a number of chemical agents and biochemical processes.

The principal method studied was that of eutoxidation. (Avery ,
McLeod, and McCarty, and others) It was found that ascorbic acifi
was effective in inactivating the desoxvribonucleic acid cf the
pneumococcal transformation svstem. The addition of substances
containing sulfhydryl groups was observed to inhibit the inactiv-
ating pronertv of the ascorbic acid. This mav have been due to
the oxidation of sulfhvdrvl groups, -SH, to the disulfide link-
age, -S-S-. In other words a transfer of the activity of the
ascorbic acid oxidizing mechanism was effected fro one substratlle
to another, perhaps to one more readily capable of oxidation. It
was further observed that the enzyme catalase also inhibited the
action of ascorbic acid, when hvdrogen peroxide was added to a
preparation of desoxyribonucleic acid, ascorbic acid, and catal-
ase, the desoxvribonucleic acid was inactivated. Other peroxidef
were also effective in inactivating the transforming substance.

51

but it was seen that none of these, including hydrogen peroxide,
were as effective as inhibitors as was the ascorbic acid. The
investigators concluded that the mechanism of inhibition was one
of autoxidation . Cupric ion, which is known to increase the rate
of autoxidation of ascorbic acia, when added to the inhibiting
system, not bly increasea the effectiveness of the ascorbic acid
as an inactivator of desoxyribonucleic acid, even when the cupric
ion was present to the extent of 0.00001 mole oer liter.

It is extremely interesting to note here that the pneumo-
coccus elaborates not only a serologically specific nucleic acid
of t e desox rribose ty e, but also catalase, hydrogen neroxide,
and desoxyribonuclease . These sub tances, the possible inter-
action of v.'hich may readily be seen, serve to illustrate, in a
minute way, the complexities of the physiological and biochemic-
al process es of bacterial cells. The four substances may be sep-
arated into two systems, each consisting of an enzyme and its
substrate, the one, hydrogen peroxide am catalase, and the othe:*
desoxyribonucleic acid and desoxyribonuclease. These two systems
are linked together bv the fact that the hydrogen peroxide acts
as an inhibitor of the desoxvribonucleic acid.

These four substances are of interest in connection vvrith
their roles in the dissociation of 3 to R forms and their rever-
sion. It was noted (Avery, McLeod, and McCarty) that serum as a
component of the culture medium used in the transforming system
is essential. It would therefore seem reasonable to conclude
that there are in normal serum one or more substances which have
an important role in the process of transformation. ; hen, for

r:

-

5.2

example, Type III S pneumococci are grown in normal serum-con-
taining culture media for a long time, through many transplants,
they are seen to lose the property of elaborating the capsular
polysaccharide. There may therefore be assumed to be a lack of
desoxyribonucleic acid formation. This is found to be so. The
mechanism may be either the inactivation of the original amount
present by desoxyribonuclease or by the hydrogen peroxide. How-
ever it is known that regardless of its phase of dissociation
the pneumococcus elaborates the enzyme catalases This would un-
doubtedly ue troy the hydrogen peroxide formed by the organism,
for, because of its specificity and biochemical activity it has
no other substrate u on which it may act. This apparently leaves
the desoxyribonuclease-desoxyribonucleic acid enzyme-substrate
system. It has been noted (Diehl and others) that many of the
enzymes which bacterial species elaborate are adaptive, i.e.,
they are produced by the organism only when a specific substrate

is present; the substrate may b*? only a simple compound contain-

9 H

mg a particular linkage, such as -C-N-, or it may be a complex
organic compound such as a protein or a carbohydrate. A search
of the literature has failed to reveal any data as to whether or
not desoxyribonuclease is present in cultures of non-tyne-specif ••
ic R pneumococci, such cultures being known to contain a desoxy-
ribonucleic acid fraction which is serologically inactive, and
possesses no transforming properties. It would be of interest
to know if such a situation exists.

It was found (Avery, McLeod, and McCarty) that when Type II
k pneumococci had been transformed to Type III S pneumococci

,

.

t • •

53

after treatment with the purified activating substance extracted
from cultures of Type III S pneumococci, the transforming sub-
stance is transmitted in series and can subsequently be recov-
ered in amounts far in excess of that originally used as inocul-
um. This phenomenon is strikingly analogous to that exhibited
by viruses and genes. Many viruses are known to be pure protein
molecules, and genes also are thought to be complex chemical
entities closely related to the nucleouroteins . The work of
Caspersson and others has pointed to the fact that chromosomes
and genes are apparently mixtures of nucleic acids of the ribose
and desoxvribose types. There apoears to be a balance between
these two tvpes of substances set up bv the individual cell
according to its state of biological and physiological activity.

It is difficult to say at this point exactly what is the
relationship existing between the phenomena of bacterial varia-
tions and cell mutations on the one hand, and the role of the
nucleoproteins in the Phenomena of cneumococcal transformation
and the self -duplication of viruses and genes in successive gen-
erations on the other. It is known, for example, that certain
unknown substances apparently present in serum are necessary in
the pneumococcal transforming system. The data so far obtained
are interpreted as indicating that the serum factors act by
altering the su rface of the bacterial cell so that the specific
desoxyribonucleic acid is taken up or absorbed. This PEoblem is
still in the process of being worked out. The mechanism of the
action of the transforming substance upon the metabolic processes
of the pneumococcus is yet to be discovered. It is known that

54

>

nucleoproteins of both the ribose and desoxvribose types play
important roles in the transmission of hereditary characteristic^
by way of the chromosomes and genes present in the sex cells of
plants and animals , as shown in numerous experiments performed
with the giant fruit flv .drosophila. But whereas the mutations
in the case of .drosophila have been effected by the apolication
of foreign chemical comoounds, particularly of the hydrocarbon
group, as Dhenanthrene , pneumococcal transformation has been
brought about by the application of a substance known to be nor-
mally present, not only in the cells of the pneumococcal species
but also in all other bacterial species and in all plant and
animal cells.

The problem of the presence or absence of a nucleus in bac-
terial species is still an unanswered one, but there is hope
that with the aid of the electron microscope, together with
spectrometric and physico-chemical techniques, we are at least
coming closer to an experimental solution to this problem. It
has been observed that a wide variety of bacterial species give
a positive Feulgen nucleal reaction, but this technique is still
under much controversy. Careful utilization of the nucleal reac
tion has convinced many authors that bacteria contain well-defim
Feulgen positive structures hich divide before cellular divi-
sion, which react like chromatin with the basic dyes, and which
are comparable to the chromosomes of plant and animal cells.

These structures have been termed nucleoids, chromatinic bodies,
or chromosomes. They occur in s >ores as well as in vegetative
forms, and are believed to contain the hereditary mechanism of

.

,

55

I

the bacterial cell. As pointed out previously in the introduc-
tion to this caper, the determination of the presence or absence
of a nucleus, and further of a process of nuclear division, or
possibly of chromosomic division would be a great advance in the
study of the rocesses involved in karyokinesis and in the trans •
mission in series of chromosomes and genes from one generation
to the next. For we pointed out before that in the case of a
bacterial scecies, e.g., Diplococcus pneumoniae, we have an ent-
ire evolutionary process or cycle before us in the short soace
of two or three days, a single cell producing 4 0 to 50 genera-
tions, or about 5X10'*'§rganisms , within 72 hours. It would re-
quire about 1000 years to produce this number of organisms in
the species homo sapiens , provided they all survived.

Since the different components and properties of the bacter-
ial cell can vary independently of one another, it is possible
to obtain a large number of variant forms which, as we have
repeatedly emphasized, can be used as reagents in the analysis
of bacteriological phenomena. The comparative study of the dif-
ferent variants of one given culture has so far been largely
limited to their morphological structures, but there is little
doubt that it could apply also to their biochemical processes.

It w ould be interesting to know, for instance, whether the ab-
sence of the type-specific polysaccharides or proteins in certaij(i
variants of the salmonella^ pneumococci streptococci, etc., is
due to the fact that these substances are not produced, or to
the fact that t hev are metabolized further, and thus cannot ac-
cumulate .

. • X

56

i

This is indeed the most striking phenomenon revealed by the
study of bacterial variability. The cell can live successfully
and continue its existence and multiply as an independent living
object after having lost a great variety of structures and func-
tions which had appeared to constitute important components and
attributes of the ’’normal" parent form. These structures and
functions can be lost and regained indenendentlv of one another,
without altering the essential nature of the gerxi , or the potent-
ialities of the cell. It is even possible to substitute experi-
mentallv one character for another, to cause, for instance, a
strain of pneumococcus to produce, and to transfer to its pro-
geny the ability to produce, a polysaccharide different from the
one it has been known to synthesize heretofore. Not only does
the cell appear as an integrated complex of independent charac-
ters, but it is possible to substitute for one of these charac-
ters another one homologous, but different, without interfering
essentially with cellular organization.

All living objects, whatever their nature or dimensions,
obev the same natural laws; it is not doubted that the study of
bacteria, like that of other cells, will progress v ith the under-
standing of the physic ochhmical phenomena which are the mani-
festations of their living processes. But each science has, in
addition to that fund common to all departments of knowledge,
its particular genius determined by the peculiarities of the
material which it studies. The extraordinary plasticity of
bacteria, the ease with which they adapt themselves to the en-
vironment, has not only determined their importance in the eco-

homy of nature; it also makes them ideal objects for the study
of that organization and integration of independent characters
which define and characterize life.

.

ABSTRACT

5B

The stud:'- of bacteriological phenomena is the study of many
millions of organisms acting simultaneous 137-. It is known that
mutations occur naturally in 1 out of lO5 or 10 cell divisions,
and though this natural frequency cannot be increased, the sur-
vival rate of the mutant form can be increased by various arti-
ficial cultural methods. By studying these mutations and the
effects that may be produced upon their biochemical, physiolog-
ical, serological, structural, antigenic and morphological mani-
festations and processes, we may eventually approach the solu-
tion of some of the Problems of medical bacteriology in the
fields of epidemiology, histolog3r, and cytochemical studies of
higher plant and animal cells.

Bacteria show marked pleomorphism even within a single bac-
terial species. These variations are due to a number of differ-
ent factors; in the case of the bacillary forms there appears to
be an internal axially disposed force which, depending upon the
surface tension of the surrounding medium, tends to counteract
the rounding effect of that tension u on the cells, thus deter-
mining the shape of the cells. The period of growth has a deter
minative effect, the cells usually being larger or longer in the
logarithmic phase of growth. This last factor is also interde-
pendent upon the factor of bacterial nutrition, and upon certain
substances or factors that may be present in culture media. For
examnle, lanceolate! diplococci of Type III S pneumococcus grown
for a long time in Type III S antiserum will exhibit long rods

.

.

.

'

59

and filamentous f orrns . Practically all bacteria exhibit slight
morphological variations during their normal growth curve,
though some may show such marked variations as in their normal
grouping, the loss of certain structures, such as capsules and
flagella, and variations in the internal structure of the cell.

Colonial variation within a species indicates the occurrence
of mutant forms. Colonial morphology is largely dependent upon
the structure of the organisms comprising the colony, and to a
lesser dgree uuop their biochemical and Physiological properties
Colonial organization can be related to the method of growth of
the organism, by correlating it with three types of post-fission*-
al movement, snapping, whipping , and sliding. It has been ob-
served that all bacterial species exhibit several types of
morphologically distinct colonial forms. These are stabilized
through successive generations towards one phase or another.
Colonial morphology may be associated with other bacterial pro-
perties, such as virulence, encapsulation, enf lagellation , and
motility. In the case of certain species, colonial morphology
may be associated with variations in serological and biochemical
manifestations. Three distinct colonial types may be described.
The M, or mucoid consists of virulent, encapsulated organisms.

The S, or smooth consists of non-capsulated, possibly flagellate 1
organisms which, depending upon the species are virulent or a-
virulent. The R, or rough consists of non-capsulated, non-
flagellated, avirulent organisms. These rules are not hard and
fast, for we have the S and R forms of Klebsiella pneumoniae
that are encapsulated, and the smooth non-motile and rough mot-

.

<

"

*

60

tile variants of the Colon-Salmonella organisms.

The variations of the pneumococcus which were discussed wei
those of colonial formation, immunochemical specificity, and the
transformation of pneumococcal types. A brief historical outlir
of the studies of the pneumococcus has been given, including the
work of Collins, Kiss, and Park and Williams, the serological
classification of the organisms bv Weufeld and Levinthal and the
transformation of types in vivo and in vitro , which we re effecte
bv Griffith, Alloway, Dawson and Sia, and Avery, McLeod and
McCartv. Three phases of colonial dissociation were found to be
associated with the Pneumococcus. The M, or mucoid Phase con-
sists of encapsulated, virulent organisms showing tvpe specific
itv. The S, or smooth phase consists of non-cansulated , avirul-
ent organisms, showing species specificity, and retaining the
property of reverting to the mucoid chase. The R, or rough
chase consists of non-cansulated, avirulent organisms, exhibit-
ing marked pleomornhism, showing Ion-' rods and filamentous forms
often negative to the gram stain, and no longer capable of re-
verting to the smooth or mucoid phases. The rough chase was
also shown to be insoluble in bile, and in some cases it was
incaoable of fermenting inulin.

The work on the transformation of pneumococcal types is
based upon the ex erimental data set forth in 1917 bv Dochez and
Avery, that each pneumococcal tvpe elaborates a specific soluble
substance, i.e., its capsular polysaccharide, which is peculiar
only to that ty e. These substances were isolated in pure form
from cultural extracts of pneumococci, and analyzed by a number

.

61

of workers who observe,, that while each of these substances gave
positive oreci itin tests with their res active ant ipneunococca]
sera, thev were of themselves non-antigenic . Seventh-five tv es
of ce sular polvsaccharides have been differentiated. Certain
other variations of the nneuraococcus have been observed to be
associated with variations in capsular elaboration, namely, re-
sist nee to -hagocTrtosis , virulence, tvne specificity, and color-
ial morphology. Griffith found in 1923 that R forms of the
pneumococcus are identical in all their iroaerties, regardless
of the 3 tyne from v/hich thev were derived, and dependent upon
the extent of dissociation. When these R forms were passed
through mice, or were grown in ani-R serum, S forms v/ere ob-
taine v/hose tyne specificitv was that of the S organisms from
which the R form was derived, and never of any other tvne.
Allowav reported the same results when R forms v/ere grown in
immune anti-R serum. He further found that R forms of one type
grown in cell-free extracts of a heterologous S type were trans-
forme to S forms of the t^pe specificitv of the cell-free ex-
tract. This type specificity was seen to be transmissible in
series .

In 194-4- Avery, McLeod and McCarty isolated the active
transforming substance from cultures of Type III S pneumococci.
This substance w as analyzed and found to be a nucleic acid of
the desoxyribose tvpe . Chemical, phvsical and biological tests
were used to determine the purity of the activating substance.
Several substances, notably those v/hich v/ere able to set up
autoxidative svstems, such as ascorbic acid, v/ere found to inac-

■

.

62

tivate the transforming substance. These inactivating reactions
were found to be reversible, except when the desoxyribonucleic
acid was inactivated by treatment with desoxyribonuc lease , an
enzyme isolated from beef pancreas, which irreversibly destroys
the nucleic acid. The pneumococcus was observer to elabor- te
not only the inducing substsrce, but also several of the inac-
tivating substances. Several factors in serum were found to be
ecwential to the transforming system. The role of enzymes as
biological tools in histological and cvtochemical studies has
been nointed out. The transformation of pneumoco' cal types "nd
the subsenuent transmission of the new hereditary unit has been
comnared histologically and physico-chemicallv v.7ith the self-
duplication of viruses end genes. The problem of a nucleus or
of nuclear material in bacteria is discussed briefly in connec-
tion with the substances involved in the transformation of
pneumococcal tvpes.

►

I * *

63

BIBLIOGRAPHY

Alloway , J.L.: The transformation in vitro of R pneumococci into
8 forms of different specific tjrpes by the use of filtered
pneumococcus extracts. J. Exp. Med. ,1932,33 , 91-99
Alloway, J.L.: Further observations on the use of pneumococcus
extracts in effecting transformation of tyre in vitro.

J. Exp. Mea. ,1933,57,265-278

Amoss, H.L . : Specific soluble substances of the pneumococcus in
the blood in pneumonia. Pros .Soc .Exp. Biol .Med . ,1930,26,23
Andrewes, F.W.: Studies in group agglutination. I. The Salmonella
group and its antigenic structure. J. Path. &Bact ., 1920 , 23
503-521

Arkwright, J.A.: Variation in bacteria in relation to agglutina-
tion by salts and by specific sera. J.Path.&Baot . ,1922, 25 ,
358-360

Arkwright, J.A. : Variation in bacteria in relation to agglutina-
tion by salts ana by specific serum. J.Path.&Bact. ,1921,24,
36-60

Arkwright, J.A., and Pitt, R.M. : The effect of growing smooth

and rough cultures in serum. J.Path.&Bact. ,1829. 32,2^9-246
Avery, O.T.: A further study on the biologic classification of
pneumococci. J. Exp. Med. ,1915 , 22,804-819
Avery, O.T.: The role of specific carbohydrates in pneumococcus
infection and immunity. ANh. Intern. Med . ,1932-1933.6 ,1-9
Avery, O.T. , and Cullen, G.E.: Studies of the enzymes of the
pneumococcus. J. Exp. Med. ,1920,32,547-593
Avery, O.T., and Dubos , R. J. : The protective action of a specific
enzyme against Type III Pneumococcus infection in mice.

J. Exp. Med. ,1931,54,73-89

Avery, O.T., and Goebel, W.F.: Chemo-iramunological studies on

the soluble specific substance of the pneumococcus. I. The
isolation and properties of the acetyl polysaccharide of
pneum coccus Type I. J. Exp. lied. ,1933 , 38, 731-755
Avery, O.T., and Heidelberger , M. : Immunological relationships
of cell constituents of pneumococcus. J. Exp. Med . ,1925 ,42 ,
367-376

Avery, O.T., Heidelberger, M. , and Goebel, W.F.: The soluble

specific substance of Friec-lKnder fs bacillus. II. Chemical
and immunological relationships of Pneumococcus Type II
and of a strain of Friedlander ’ s bacillus. J .Exp. Med. , 1925 ,
42,709-725

Avery, O.T., MacLeod, C.M., and McCarty, M. : Studies on the

chemical nature of the substance inducing transformation of
pneumococcal types. J. Exp. Med. ,1944,79,137-158
Avery, O.T., and Morgan, H.J.: Immunological reactions of the
isolated carbohydrate and rotein of the pneumococcus.

J. Exp. Med. ,1925.42.347-353

Avery, O.T., and Neill, J.M.: The antigenic properties of solu-
tions of pneumococcus. J. Exp. Ivied. ,1925 ,42,355-365
Avery, R.C., and Leland, S.J.: A simple method for the isolation
of pure cultures from single bacterial cells. J. Exp. Med.,
1927,45,1003-1007

Bartholomew, J.W. ; and Umbreit, W.W.: Ribonucleic acid and the
gram stain. J.Bact. ,194-4,47,41$

Welding, D.L. , and Marston, A.T.: Medical Bacteriology. Appleton-

>

Centura, Co . . New York. 1938.

Bergev, D.H. , Breed, R.S., Murray, B.G. , and Hitchins , A.P.:

Bergey’s Manual of Determinative Bacteriology. 5th Edition,
1939. Williams and Wilkins Co. Baltimore
Berrv, G.P. , and Dedrick, H.M. : A method for changing the virus
of rabbit f ibroma^Shope ) into th t if infectious myxomat-
osis (Sanarelli ) . J.Bact. ,1936,31,50-51
Bisset, K.A. : The structure of rough and smooth colonies.

J.Bact. , 1938,47,223-229

Bisset, K.A. : The structure and mode of growth of bacterial col-
onies morphologically intermediate between R and S forms.
J.Path ,&Bsct . ,1939,49,491-496

Bisset, K.A. : The mode of growth of bacterial colonies. J.Path.
MBact . ,1939,48,427

Blake, F.G.: Methods for determination of pneumococcus types.

J. Exp. Med. ,1917,26,67-80

Blake, F.G., and Trask, J.D.: Pneumococcus variants intermediate
between 3 and R forms. J.Bact . ,1933.25.289

Boyd, «M. C.: Fundamentals of Immunology. Interscience Publisher 3
Inc., New York, 1943

Brown, R. : Chemical and immunological studies of the meumo-

coccus. V. The soluble specific substances of Tvpes I -XXXII.
J .Immunol . .1939. 37.445-455

Bruner, -j.W. , and Edwards, P.R.: A note on the monphasic non-
specific salmonella types. J.Bact. ,1939,37,365-370
Bruner, D.W. , and Edwards, P.R. : The demonstration of non-spec-
ific components in Salmonella paratv 'hi A bv induced varia-
tion. J.Bact. ,1941.42,476-478

Clough, M.C.: An observation of a mutation of one of the strains
of pneumococcus. J. Exp. Med. ,1919.30.123-147
Coli.ins, K.R. : The application of the reaction of agglutination
to the pneumococcus. J. Exp. Med. ,1905.7,420-429
Cooper, G. , Edwards, k. , and Rosenstein, C.: The separation of
t:roes among the pneumococci hitherto called Group IV and
the development of the therapeutic antiserums for these
types. J. Exp. Med. .1929.49.461-437
Daw on, M.H.: The interconvertibility of "R” and "S" forms of
pneumococcus. J. Exp. Med. ,1928,17.577-591
Dawson, M.H.; The transformation of neumococcal types. I. The
conversion of R forms of pneumococcus into S forms of the
homologous type. J. Exp. Mel . , 1929,51 , 99-122
Dawson, M.H. : The transformation of pneumococcal types. II. The
interconvertibility of type-specific S pneumococci.

J. Exp. Med. .1930.51.123-147

Daw on, K.H.: Transformation and dissociation of pneumococcus.

J. Clin. Invest . .1933 , 12 , 978-10 jJD
^awson, M.H. : Dissociation of the pneumoc ecus; a new colony
variant. Proc .Soc .Exp .Biol .Med . , 1933 . 30 . 806-808
Dawson, M.H. : Variations in the Pneumococcus* J . . : . Bact.,

1934,39,323-344

64

: . . , ! , . . . -30

-

• . .

. .

: . . . . , , . .

... , '

.

.

.

. .

. .

.

- *

.

i • •

. . • < Vile

. ' ,

.

: •

.

.

t ■ ? •

: . . ...

I

. .

t •

: . . ,

. . . .

. • 1 '-'l ' r ' ■ . : •

f

.

• *

. .

.

.

t t. .

, . . .

. . ,

.

, .

.

65

Dawson, M.H. , and Avery, O.T.: Reversion of avirulent "rough"
forms of pneumococcus to virulent 'smooth'1 types. Eroc .

Soc .i^xp. Biol .lied. ,1927 , 24,943

i)awson, M.H. , Hobby, G.L., and Omnstead, k. : Variations in the
hemolytic streptococci. J. Infect. Jis . ,1936 162, 13 8-168
Dawson, M.H. , and Sia, R.H.P. : In vitro transformation of

pneumococcus types. I. A technique for inducing transform-
ation of pneumococcal tv >es in vitro. J .Exp. Med , ,1931 .54-1
681-699

De Kruif , P.H. : Mutations of the bacillus of rabbit septicemia.
J. Exp. Med. ,1922,35,561-574

De Kruif, P.H. : Virulence and mutation of the bacillus of rabbit
seoticemia. J. Ex u. Lied. ,1922 , 55 ,621-633
Deskowitz, M.W. : Bacterial variation as studied ir certain un-
stable variants. J.Bact. ,1937* 33 >34-9-367
Diehl, H.S.: The specificity of bacteriolytic enzymes and their
formation. J. Inf .Pis. ,1919.24,347-361
Diere, C.J.: On the "activation ,T of the lactase of Escherichia
coli mutabile . J.Bact . ,1939, 37,473-483
Dochez, A.R., and Avery, O.T.: The elaboration of specific solu-
ble substance bv pneumococcus during growth. J. Exp. Med. ,
1917 , 26 , 477-493

Dochez, A.R., and Gillespie, L. J. : A biologic classification of
pneumococci bv means of immunity reactions. I. A. M. A. , 1913
61,727

Dubos , R.J.: The Bacterial Cell. Harvard University Press.
Cambridge. 1945

Dubos, R.J.: Eactors affecting the yield of specific enzyme in
cultures of the bacillus decomposing the capsular poly-
saccharide of Type III pneumococcus. J. Exp. Med. .1932.55 .
377-391

Dubos, R.J.: The autolvtic system of oneumococci. J. Exp. Med. ,
1937,0,05,873-883

Dubos, R.J.: Mechanism of the lysis of pneumococci bv freezing
and thawing, bile and other agents. J .Exp .Med. ,1937, d ,66 ,
101-112

Dubos, R.J.: The effect of the bacteriolytic enzyme of pneumo-
coccus on the antigenicity of the encapsulated pneumococci.
J.Ex - ,Meu. , 1937, d, 66, 113-123
uubos , R.J.: The effect of formaldehyde on oneumococci.

J. Exp. Med. ,1938, a, 67, 389-398

Dubos, R.J.: The ad ptive production of enzymes by bacteria.

Bact .Rev. ,1940.4,1-16

Dubos, R.J. a,d Avery, O.T.: Decomposition of the capsular poly

saccharide of pneumococcus Type III bv a bacterial enzyme .
J. Exp .Med, ,1931,54,51-71

Eaton, u.D.: Studies on pneumococcus variation. I. Variants char-
acterized by rapid lysis and absence of normal growth under
the routine method of cultivation. J.Bact. ,1934,27,271-291
Eaton, M.D. : Studies on pneumococcus vari tion. II. Smooth viru-i
lent variants produced by daughter-colony dissociation of smooth
pneumococcus strains. J.Bact . , 1935 , 30,119-135

.

• •

.

*

•

.

.

9

.

• i

.

. .

• • •

• f • -

•

: . .

• •

t

.

t •

.

t « « -•

.

' t

c

(

65

tinder, J.F., and Wu, C.J.: An immunological study of the A sub-
stance or acetyl polysaccharide of Pneumococcus Type I.

J. Exp. Med. ,1934,60,127-147

Felton, L.D.: Studies on virulence. IV. Influence on virulence
of pneumococci of growth on various me.ia. J. Exp .Med. ,
1932,56,13-23

Fletcher, W. : Capsulate mucoid forms of paratyphiod and dysen-
tery bacilli. Lancet , 1916,11 , 102-104

Gause, G.F.: Some physiological properties of dextral and sini-
stral forms in Bacillus mycoides. Biol .Bull. , 1939 , 76 , 448-
465

Gilbert, I.: Jissociation in an encapsulated staohvlococcus .
J.Bact. ,1931,21,157-160

Goebel, W.F.: The preparation of the type-specific polysacchar-
ides of pneumococcus. J.Biol.Chem. ,19301691395-416

Goebel, W.F.: Chemo -immunological studies on the specific solu-
ble substance of pneumococcus. II. The chemical basis for
the immunological relationship between the capsular poly-
saccharides of Types III and VIII pneumococcus. J.Biol.

Chem, 1935, 110, 391-398

Goebel, W.F.: Ghemo-immunological studies on conjugated carbo-
hydrate-proteins. XII. The imiduno logical properties of an
artificial antigen containing cellbiuronic acid.

J. Exp .Med. ,1940,72,33-48

Goebel, W.F., and Averv, O.T.: A study of pneumococcal autolysis.
J.Ex i.Mec- . ,1929,49,267-286.

Goebel, V/.F., and Avery, O.T.: Ghemo-immunological studies on
conjugated carbohvdrate-pr teins. J. Exp. Med . ,1931 , 54, 431-
456

Greenstein, J.P.: Nucleooroteins . Adv. in Prot.Chem. ,1944,1,
209-281

Greenstein, J.P., and lenrette, W.V. : Physical changes induced
in thymo nucleic acid by proteins, salts, tissue extracts,
and ultraviolet irradiation. Cold Spring Harbor Symposia
4uant. Biol. ,1941,9,236-252

Griffith, F.: The influence of immune serum on the biological
procerties of pneumococci. Report 18 on Public Health
and Medical Subjects .Ministry of Health, 1923 , op • 1-13

Griffith, F . : The significance of pneumococcal types. J .Hyg. ,
1928,27,113-159

Hadley, P. : Microbic dissociation. J. Infect .Pis . ,1927,40,1-312

Hadley, P. : Further advances in the study of microbic dissocia-
tion. I, Inf ect .Pis . ,1937,60,129-192

Hadley, P. : Bearing of dissociative variation on the species-
concept among the schizomycetes . J.Inf ect .Dis . , 1939, b ,65 ,
267-272

Heidelberger , M. : Immunologic ally specific polysaccharides.

Chem. Rev. ,1927,3,403-425

Heidelberger, H. : The chemical nature of immune substances.
Physiol. Rev. ,1927,7,107-128

Heidelberger, M. , and Avery, O.T. : Soluble specific substance
of the pneumococcus. J. Exp, lied. ,1923 , 38,73-85

, .

.

t . . .

\ .

.

.

. t .

.

. : .

t • * «

: . . , i. •.

,

.

t . . <

. .

.

.

• . . .

, . . .

•

: •

•

.

• »

■

■ e

? . «

.

1 »

.

.

: .
.

.

t .

. .

.

66

Heidelberger, M. , and Avery, O.T.: The soluble specific substance
of pneumococcus. J. Exp. Med. , 1924 , 40 » 301-316
Heidelberger, M. , Goebel, I.F., and Avery, O.T.: The soluble
specific substance of the pneumococcus . J. Exp. Med . , 1925 ,
42,727-745

Heidelberger, 41., Kabat, 3. A. , and Meyer, M. : A further study of
the cross-reactions between the specific polysaccharides
of Types III and VIII pneumococci in horse antisera.

J. Exp. Med. , 1942 ,75,35-48

Heidelberger, El, Kendall, F.E., and Bchero, H.W.: Specific poly-
saccharides of Type I, II, and III pneumococcus; revision
of methods and data. J. Exp. Med. ,1936,64,559-572
Heidelberger, M. , and Schero, H.W. : Protein fraction of a strain
of Group MA,T hemolytic streptococci. J. Immunol. ,1934,37,
563-581

Hiss, .H. : A comparative studi of the pneumococci and allied
organisms. J. Exp. Med. ,1905,7,547-582
Hitchcock, C.H.: Serological relationships among and between
the streptococci and pneumoco ci. J. Exp. Med. , 1925 , 41 ,

13-19

Hobby, G.L. , and Dawson, M.H.: The encapsulation of hemolytic
streptococci. Brit. J. Ex .Path. ,1937,18,212-214
Hoogerheide, J.C.: Studies on capsule formation. I. The condi-
tions under which Klebsiella pneumoniae forms capsules.

I. Bact. ,1939,38,367-388

Hoogerheide, J.C.: Studies o. capsule formation. II. The in-
fluence of electrolytes on capsule formation by Klebsiella
pneumoniae . J.Bact . ,1940, 39,649-658
Huntoon, F.M.: Antibodv studies. III. Chemical nature of anti-
body. J. Immunol . , 1921 ,6 , 185-200
Julianelle, L.A.: A biological classification of Encapsulatus
pn umoniae (Friedlflnder *s bacillus). J. Exp. Med. ,1926, a, 44,
113-123

Julianelle, L.A.: Immunological elationships of encapsulated

and capsule-free strains of encapsulatus pneumoniae (Eried-
l&nder’s bacillus). J. Ext .Med. ,1926 ,b,44,683— Mm ,735-751
Julianelle, L.A.: Bacterial variation in cultures of FriealSnd-
er’s bacillus. J. Exp. I led . , 1928 , 7 , C 9-902

Julianelle, L.A.: Immunolog cal specificity of Bacterium aero-
genes and its antigenic relation to pneumococcus Type II
and Friedlfinder Ts bacillus. J. Exp. lied. ,1937,32,21-33
Julianelle, L.A.: Immunological relationships of gram negative
rods, .roc .Soc .Exp. Biol, i-ied. ,1937,36 ,245-258
Kolchin, 3.3. , and Gross, C.: Observations on the cross-protec-
tion of antipneumococcus monovalent sera Type I, II, and III

J. Immunol. ,1924,9, 505-513

Hinton, R.W. , Beal, S.C.; and Mitra , B.N.: Chemical and sero-
logical variation in single-cell cultures of Vibrio cholera 3
and related organisms. Indian J. Med. Res . ,1938,25 ,575-5^4
MacLeod, C .M. : Determination of tvpes of Corvnebacterium dich-
theriae . Bact.Rev. ,1946,7,1-41

. . . ,• -
. t . ...

. . . . . t ... t •• •

.

, < . . , . .

.

.

' t . . . .

: . . t

t . . . .

. .

• •

? •

:

, ,

.

. .

: . .

•

, . ,

. .

. . f

.

.

r •

r . .

. • •

» _ . » t

« •

.

.

. •
.

: . .

. . ,

t ■ .

. .

. . ,

... ...

: . , .

«... .. ..

t. ’

67

McCarty, M. : Reversible inactivation of the substance inducing
transformation of neumococcal types. J.Lxp. Med. ,1945,81

501-5U

McCarty, M. : Purification and properties of desoxvribonuclease
isolated from beef pancreas. J. Gen .Physiol . ,1946 , 29 , 123-
139

McCarty, M. : Chemical nature and biological specificity of the
substance inducing transformation of pneumococcal types.

Bact .Rev. ,1946 , 10 .63-71

McCarty, M. , and Avery, O.T.: Studies on the chemical nature of
the substance inducing transformation of pneumococcal
tyoes. J.exp .Lieu . , 194b , 83 , 89-104

Meyer, K., Smyth, E.M. , and Dawson, M.H.: The nature of the muco-
polysaccharides of synovial fluid. Science, 1938*68, 129-147
Mirsky , A.E.: Chromosomes and nucleo roteins . Adv , in Lnzymologv ,
1943,3,1-34

Morgan, H. ., and Beckwith, T. J. : Mucoid dissociation in the

colon-typhiid-salmonella group. J .Inf .Pis . ,1939,63 , 113 -124
Morton, H.E.: Corynebacterium diphtherias. A correlation of
variations within the species. Bact. Rev. , 1940 , 4, 177-226
Mudd, S., Heinmets , F., and Anderson, T.F.: Bacterial morphology
as shown by the electron microscope. VI. Capsule, cell-wall,
and in er roto lasm of pneumococcus Type III. J.Bact . ,
1943,46,205-211

Mudd, S., Heinmets, F., and Anderson, T.F.: The pneumo-
coccal capsular swelling reaction, studied with the aid of
the electron micr >e. J.Lxp. Mo -.,1943,76.327-332
Mudd, S., and seiner, M. : The antigenic structure of hemolytic
streptococci of Lancefield group A. XI. Relationships of
the nucleo proteins of some species of streptococci and
pneumococci. J. Immunol. ,1942,45 ,21-28
Neufeld, F. , and Levinthal, W.: Beitr&ge zur variability der
oneumokokken . 4 . Imraunitatsf orsch . , 1928 ,53, 324-340
Niell, J.M., and Avery, O.T.: Studies on oxidation and reduction
by pneumococcus. VI. The oxidation of enzymes in sterile ex ?
tracts of pneumococcus. J. Exp. Med. ,1924,40,405-422
Nungester, W.J.: Independent variation of bacterial properties.
J.Bact. ,1933 , 25 , 49-30

Pappenheimer , A.M.,Jr., and _;nders , J.F.: Specific carbohydrate
of Type I pneumococcus. Proc .Soc .Exp. Biol .lied . , 1933 , 31 ,

37-39

Park, H.W., and Williams, A.W.: A study of pneumococci. J.Lxp.
Med. ,1905,7.403-419

Park, H.W., and Williams , A.W. : Pathogenic Microorganisms.

Lea & Febiger, Philadelphia. 1939
Paul, J.R.: A comparative studv of smooth and rough pneumococcus
colonies. J. Exp. Med. ,1927,46,793-805
Paul, J.R.: Pneumococcus variants. J.Bact. ,1934,23,45-64
Randall, v/.A. : Colonv and antigenic variations in Klebsiella
pneumoniae Types A, B, and C. J.Bact. ,1939, 38,461-477
Reed, G.B.: A hypothetical view of bacterial variation. J.Bact.,
1933 ,2£, 580-586

t ...

. ,

...

.

.

• *

m

. . . __

• • f • •

.

. ■

0

: . .

• • •

.

■

.

. .

.

.

.

.

- . , . .

f

t

Riemann, H.A. : Serological relationships of type-srecif ic and
degraded pneumococci. J .Exp .lied . , 1926 , 43 , 107
Riemann, E.A. : Studies concerning the relationship between
pneumococci and streptococci. J .Exp .lied . , 1927 ,45 , 1-10
Riemann, K.A. : Variations in specificity and virulence of pneumo
cocci during growth _in vitro . J .Exp .ivied . , 1925 , 11 » $87-600
Riemann, H.A. : The occurrence of degraded pneumococci in vivo .

J .Exp .Med. ,1927.45,807-314

Riemann, H.A. : Reversion of R to S neumococci , J. Exp .Med. ,1929.
49,237-249

Rosenow, E.G.: Transmutation within the streptococcus-pneumo-
coccus group. J. Infect .Pis . ,1914,14,1-32
Shew, M. : Decomposition of pneumococcus carbohydrate by the com-
bined activity of strains of two bacterial species.

J.Bact. ,1937, 33 ,644-645

Shinn, L.E.: The ultimate R culture phase of the pneumococcus.
J.Bact. ,1937,33,18-19

Sia, R.H.P., and Dawson, M.H. : In vitro transformation of pneumo
coccal types. II. The nature of the factor responsible for
the transformation of pneumococcal types. J.Exo.Med . ,1931 ,
34,701-710

Sickles, G.M., and Shaw, M. : a systematic study of microorgan-
isms which decomoose the specific carbohydrate of the
pneumococcus. J.Bact. ,1934,28,413-431
Sickles, G.M. , and Shaw, M. : Microorganisms which decompose the
specific carbohydrates of pneumococcus Types II and III.
Sickles, G.M. , and Shaw, M. : A microorganism which decomposes
•the specific carbohydrate of pneumococcus Type VIII.

^roc .Soc .Exp .Biol .Med. ,1933 , 32,857-858
Smith, T., and Reagh, A.L.: The non-identity of agglutenins act-
ing upon the flagella and upon the body of bacteria.

J. Med. Res . , 1903 , 10 , 89-100

Stillman, E.G.: A study of atypical Type II pneumococci.

J. Exp. Med. ,1919,29,231-258

Stryker, L.M.: Variations in the neumococcus induced by growth
ip immune serum. J. Exp. Med. ,1916,24,49-68
Sumner, J.B., and Somers, G . 7b" . : Chemistrv and Methods of Enzymes
Acedemic Press, Inc. 1943

Thomason, R.H.S., and Dub os , R. J . : The isolation of nucleic acid
and nucleooroteins from pneumococci. J. Biol .Chem, 1938,125 ,
b5-74

Todd, E.W., and Lancefield, R.C.: Variants of hemolytic strepto-
cocci; their relation to tvpe-s oecif ic substance, virulence
and toxin. J. Exp. Med. ,1928,48,751-767
Toplev, vv.1v. C., and Wilson, G.S.: The Principles of Bacteriology
and Immunity. 1VM. Wood & Co., 2nd ed. Baltimore. 1937
vvhite, P.B.: Regarding alleged transmutation in vibrios. J.Path,
&3act. ,1937,44,490-492

Yudkin, J.: Enzyme variation in bacteria. Biol .Rev, Cambridge
Philos. Soc . ,1938,13,93-106

Einsser, H. , and Bayne- Jones, S.: A Textbook of Bacteriology.

D. Applet on -Century Co. New York. 1934

»

68

- — —

• . . . .

- . , f <

t ' . . .

• : . .

t ‘ . .

- v ;

. .

,

. . . ,
.

« i .

t .n t. . ,

7 < . .

: . f , . .

.

» ... ...

: . . f ,

,

, .

. . t. . ,

: . . t , . . .

. . ,

.

....

, . T

. .

.

: . i ,

* i 1 < •

, , ,v. •